UCSF Reports on Industry Activity from Outside UCSF

Title TOBACCO AND TRANSITION: UNDERSTANDING THE IMPACT OF TRANSITION ON TOBACCO USE AND CONTROL IN THE FORMER SOVIET UNION

Permalink https://escholarship.org/uc/item/3rw2c04c

Author Anna BC Gilmore

Publication Date 2005-06-01

eScholarship.org Powered by the California Digital Library University of California

TOBACCO AND TRANSITION: UNDERSTANDING THE IMPACT OF TRANSITION ON TOBACCO USE AND CONTROL IN THE FORMER SOVIET UNION

______

Dr Anna BC Gilmore

Thesis submitted for the degree of Doctor of Philosophy of the University of London

2005

London School of Hygiene and Tropical Medicine

ABSTRACT

The collapse of the Soviet Union in 1991 and its transfer from a command to a market economy precipitated immense change changes in the region’s tobacco industry. Most notable were the rapid and unregulated entry of the multinational and transnational tobacco companies (TTCs) and the privatisation of previously state owned tobacco industries. The impact of investment liberalisation and privatisation of state owned tobacco companies on patterns of tobacco use and tobacco control has not previously been studied, timely, accurate data on smoking patterns in the former Soviet Union are scarce and there has been no formal research on TTC activities there. This thesis capitalises on the unique social experiment provided by the Soviet Union’s transformation to address these research gaps through a combination of data and methodologies: analysis of routine and ad hoc data on investments and consumption; survey data on patterns and determinants of smoking behaviour in the region; and internal tobacco industry documents. The importance of this work is underlined by the fact that tobacco industry privatisation continues to be promoted by international actors who have seemingly given no consideration to its potential impacts, and that tobacco control policies are most effective when informed by accurate data on patterns and determinants of smoking and by information on industry tactics and actions. The thesis finds that between 1992 and 2000 the TTCs made significant investments in ten countries of the region. These investments have exponentially increased cigarette production rates and fuelled cigarette consumption without favourably influencing trade figures. Smoking prevalence rates in men remain amongst the highest in the world, with rates over 60% recorded in three of the eight countries surveyed. Smoking rates in women are far lower, but their pattern appears to reflect TTC activity. Tobacco industry documents outline the major negative influence TTCs have had on tobacco control. The thesis concludes that investment liberalisation and tobacco industry privatisation pose major threats to public health and makes recommendations for action.

2 TABLE OF CONTENTS

ABSTRACT...... 2 TABLE OF CONTENTS...... 3 LIST OF TABLES ...... 8 LIST OF FIGURES...... 9 LIST OF ABBREVIATIONS AND DEFINITIONS...... 10 Abbreviations ...... 10 Definitions ...... 12 STATEMENT...... 13 ACKNOWLEDGEMENTS ...... 14 CHAPTER 1 Introduction, background, aims and objectives...... 16 SECTION A INTRODUCTION ...... 16 THESIS OUTLINE ...... 19 SECTION B BACKGROUND ...... 20 PART 1: THE FORMER SOVIET UNION...... 20 1.1 The history of the Soviet Union...... 20 1.2 The collapse of the Soviet Union...... 20 1.3 The fifteen Newly Independent States (NIS)...... 21 1.4 Economic transition ...... 23 1.4.1 The scale and speed of reform and privatisation ...... 25 1.4.2 Outcomes...... 26 1.4.3 A retrospective analysis – gathering the evidence on privatisation...... 28 1.4.4 Further evidence from CEE/FSU...... 29 1.4.5 Success or failure? ...... 30 1.5 Health impacts of transition...... 31 PART 2: THE TOBACCO INDUSTRY...... 37 2.1 The Soviet Tobacco industry ...... 37 2.2 Transition and entry of the tobacco transnationals...... 39 2.3 The history of the international tobacco industry and its global expansion ...... 42 2.4 What can be learnt from the TTC’s previous expansion to new markets?...... 44 2.4.1 TTC expansion to Latin America ...... 45 2.4.2 TTC expanstion to Asia ...... 46 2.4.3 Impacts of TTC entry to Latin America and Asia...... 48 2.4.5 Globalisation of production ...... 50 2.5 The global shift in tobacco related disease patterns...... 53 PART 3: SMOKING IN THE FSU ...... 54 3.1 Measuring tobacco use: tobacco consumption and smoking prevalence ...... 54 3.2 Consumption and smoking habits...... 54 3.3 Smoking Prevalence – adults ...... 54 3.4 Smoking prevalence - youth ...... 78 3.5 Smoking amongst doctors...... 84 PART 4: THE HEALTH IMPACT OF TOBACCO ...... 85 4.1 Chronic disease...... 85 4.2 Dose response ...... 87 4.3 Delay in health impact ...... 87 4.4 The tobacco epidemic ...... 88 4.5 Impacts of tobacco in non-western populations...... 91 4.6 Health impacts of tobacco in the FSU ...... 93 4.7 Gender differences in mortality ...... 108 4.8 Lifestyles and health promotion in the USSR/FSU ...... 109 SECTION C: AIMS AND OBJECTIVES...... 112 5.1 Gaps in knowledge...... 112 5.2 Aims...... 113 5.3 Objectives ...... 113 5.4 The research questions...... 114

3 5.5 Datasets and Methods ...... 115 CHAPTER 2 An overview of industry investments, impact and influence ...... 119 BACKGROUND...... 119 METHODS...... 120 RESULTS...... 120 Tobacco industry investments ...... 120 Contribution to FDI ...... 122 Consequences of TTC Investments ...... 132 DISCUSSION...... 135 CHAPTER 3 Exploring the impact of foreign direct investment on tobacco consumption in the former Soviet Union...... 140 BACKGROUND...... 140 METHODS...... 141 RESULTS...... 144 Tobacco leaf production ...... 144 Cigarette production ...... 147 Imports and exports ...... 148 Cigarette consumption...... 151 DISCUSSION...... 154 CHAPTER 4 Prevalence and determinants of smoking in eight countries of the former Soviet Union ...... 160 BACKGROUND...... 160 METHODS...... 161 The LLH Project...... 161 Study population and sampling procedures ...... 161 Questionnaire design ...... 162 Questionnaire administration...... 162 Statistical analyses...... 163 Analysis of smoking prevalence rates ...... 163 Analysis of smoking determinants...... 163 RESULTS...... 164 Response Rates...... 164 Sample characteristics and representativeness...... 164 Smoking prevalence ...... 166 Age at initiation ...... 171 Amount smoked and nicotine dependence ...... 171 Determinants of smoking...... 171 DISCUSSION...... 179 Strengths of these surveys ...... 179 Study limitations...... 180 Findings ...... 181 CONCLUSIONS ...... 186 CHAPTER 5 Tobacco document provenance and methodology ...... 187 BACKGROUND...... 187 The provenance of the tobacco industry documents...... 187 Access to the tobacco industry documents: Guildford vs Minnesota ...... 188 Improving access ...... 190 Use of tobacco industry documents...... 190 METHODOLOGY ...... 191 Documents used...... 191 Approach taken...... 191 The idiosyncrasies of the Guildford Archive...... 192 Understanding the provenance and weaknesses of the archive ...... 193 Planning and undertaking the search strategy...... 194 Organising the archival data ...... 195 Interpreting the data...... 197 Case studies ...... 200

4 Weaknesses...... 200 CHAPTER 6 How the transnational tobacco industry entered the former Soviet Union. Part I: Establishing cigarette imports...... 203 INTRODUCTION...... 203 METHODS...... 204 RESULTS...... 205 Reasons for BATs interest in the FSU...... 205 Preliminary steps ...... 206 Establishing BATs brands in the market ...... 207 Securing a place in the import market ...... 210 The role of smuggling in accessing new markets ...... 213 DISCUSSION...... 217 Marketing, brand and corporate identity...... 217 Smuggling...... 218 Outcomes...... 220 Summary...... 220 CHAPTER 7 How the transnational tobacco industry entered the former Soviet Union. Part II – establishing a manufacturing presence...... 222 BACKGROUND...... 222 METHODS...... 223 RESULTS...... 223 Restructuring BAT Group operations to maximise new business opportunities ...... 223 Prioritising investments within the region...... 224 Issues of competition...... 225 Tactics ...... 228 Contacts ...... 228 Selling the benefits of privatisation ...... 229 Economic arguments and excise advise...... 230 Selling BAT...... 232 DISCUSSION...... 237 BAT and its competitors...... 237 Market reform...... 238 Economic arguments ...... 238 Corporate image ...... 241 Summary...... 241 CHAPTER 8 A review of transnational tobacco company plans for a privatised tobacco industry in Moldova ...... 242 BACKGROUND...... 242 METHODS...... 243 RESULTS...... 244 The saga of tobacco industry privatisation in Moldova...... 244 Reasons for BAT’s interest...... 245 Privatisation...... 245 Marketing plans ...... 247 Leaf processing and employment ...... 249 An update...... 249 DISCUSSION...... 250 CHAPTER 9 The invisible hand: how british american tobacco precluded competition in Uzbekistan...... 253 BACKGROUND...... 253 Uzbekistan and Central Asia ...... 253 The Transition debate and corporate conduct...... 254 METHODS...... 257 RESULTS...... 258 The Uzbek Tobacco Industry...... 258 Initial Approach: committing the Uzbek authorities to negotiating exclusively with BAT...... 258 Desire to avoid a competitive tender ...... 260

5 Use of political contacts to ward off competitors and prevent a competitive tender ...... 262 Absorption of competitors ...... 264 Exclusive dealing and erecting barriers to market entry...... 264 Investment privileges...... 266 Securing the deal ...... 267 Hiccoughs at the final stages ...... 268 Outcomes...... 269 DISCUSSION...... 270 CHAPTER 10 "[A]ssisting Governments to design and implement the most suitable indirect tax systems”: BAT and tobacco taxation in Uzbekistan...... 275 INTRODUCTION...... 275 BACKGROUND...... 276 METHODS...... 277 RESULTS...... 278 BAT’s strategy for its investment in Uzbekistan...... 278 Initial focus: protective import duties...... 278 Seeking and successfully influencing contacts...... 279 Emergent obstacles...... 279 A change in approach – away from import duties to equal excise treatment...... 280 Share Purchase Agreement...... 281 Finalising the equal taxation regime...... 281 Lowering excise rates ...... 282 Achieving detailed tax reforms on ad valorem and specific tax rates...... 283 An ‘effective’ policing system...... 285 Post script ...... 286 DISCUSSION...... 287 CHAPTER 11 “Unless Health Decree 30 is amended satisfactorily it will not be possible for this transaction to proceed”: British American Tobacco’s erosion of health legislation in Uzbekistan ...... 290 BACKGROUND...... 290 METHODS...... 291 RESULTS...... 291 The marketing environment in the early 1990s ...... 291 Marketing plans ...... 292 Freedom to advertise ...... 293 Poor understanding of health issues...... 293 Health Decree 30 ...... 293 BAT’s approach...... 294 BAT’s initial Response...... 298 The initial Uzbek reaction ...... 300 BAT’s amended decree ...... 300 Escalating political pressure on the Ministry of Health...... 302 Postscript ...... 303 DISCUSSION...... 304 CHAPTER 12 Discussions and conclusions ...... 306 METHODOLOGICAL LIMITATIONS ...... 306 CONTRIBUTION TO KNOWLEDGE...... 308 What was the scale and nature of the TTC contribution to foreign direct investment in the FSU?...... 308 How have TTC investments changed the pattern of cigarette production and trade in the FSU? ...... 309 Why and how did the TTCs respond to the opportunities posed by the opening of the FSU and how did they influence the privatisation process? ...... 311 What impact has trade liberalisation and TTC entry had on tobacco consumption? ...... 315 What are the current patterns and determinants of smoking behaviour in the region and how might they be explained? ...... 317 How have the TTCs influenced tobacco control policies in the former Soviet Union and how can these influences be explained? ...... 318 Do investment liberalisation and tobacco industry privatisation pose dangers to tobacco control and health?321

6 If so, how might these dangers be mitigated?...... 323 Other lessons ...... 327 IMPLICATIONS FOR POLICY...... 327 FURTHER RESEARCH ...... 332 PERSONAL REFLECTION AND LEARNING ...... 333 REFERENCES...... 335 APPENDIX 404

7 LIST OF TABLES Table 1-1 Selected demographic and economic indicators in the FSU...... 23 Table 1-2 Trends in production of smoking materials in the USSR ...... 38 Table 1-3 Production of cigarettes as a percent of smoking material production in 1982 for selected Soviet Republics...... 38 Table 1-4 Tar and nicotine yields in USSR, US and UK cigarettes in 1986...... 38 Table 1-5 Production of filter cigarettes as a percent of total cigarette production in 1996-2001...... 41 Table 1-6 The world’s largest tobacco transnationals ...... 44 Table 1-7: Studies of smoking prevalence among adults in Armenia, Azerbaijan, Georgia and Moldova ...... 58 Table 1-8: Studies of smoking prevalence among adults in central Asia ...... 60 Table 1-9: Studies of smoking prevalence among adults in Russia...... 63 Table 1-10: Studies of smoking prevalence among adults in Ukraine...... 68 Table 1-11: Studies of smoking prevalence amongst adults in Belarus...... 71 Table 1-12: Studies of smoking prevalence among adults in Estonia...... 72 Table 1-13: Studies of smoking prevalence among adults in Latvia ...... 74 Table 1-14: Studies of smoking prevalence among adults in Lithuania ...... 76 Table 1-15 Details of youth smoking surveys conducted in the FSU...... 81 Table 1-16 HBSC data on experimental and regular youth smoking in 1993/4 (shown in brackets) and 1997/8 by sex, country and age group...... 82 Table 1-17 ESPAD data on youth smoking in 15-16 year olds (those born in 1983) in 1999...... 82 Table 1-18 GYTS data on youth smoking patterns...... 83 Table 1-19 Trends in youth smoking in Moscow ...... 83 Table 1-20 Relative risks in current versus never male smokers (except for cervical cancer in female smokers) based on UK Doctors study and CPS II. Adapted from data in Doll, Wald and Boyle ...... 86 Table 1-21 The four stages of the tobacco epidemic ...... 90 Table 1-22 The Risks of dying and proportion of deaths attributed to tobacco in men in 1990...... 97 Table 1-23 The risks of dying and proportion of deaths attributed to tobacco in women in 1990 ...... 98 Table 1-24 Standardised death rates from various causes in countries of the CIS, 1999 data*...... 98 Table 2-1 Details of tobacco industry investments in Russia ...... 123 Table 2-2 Details of tobacco industry investments in Ukraine...... 125 Table 2-3 Details of tobacco industry investments in the Baltic States ...... 126 Table 2-4 Details of tobacco industry investments in Central Asia...... 127 Table 2-5 Details of tobacco industry investments in the Caucasus...... 128 Table 2-6 Total net foreign direct investment (FDI) and tobacco investments as a proportion of total FDI. Figures in US$ millions ...... 129 Table 2-7 Percent market share for 1998-2000 (or nearest years where data available) based on legitimate sales only for all countries bar Belarus...... 130 Table 2-8 Tobacco control policies by country ...... 131 Table 3-1 Data sources and details ...... 142 Table 3-2 Cigarette consumption per capita 15+ in the Former Soviet Union (FSU) as a whole, and in countries with and without tobacco transnational investments ...... 153 Table 4-1 Characteristics of samples and countries...... 165 Table 4-2 Smoking prevalence by country, gender and age group...... 167 Table 4-3 Characteristics of smoking behaviour amongst current smokers ...... 170 Table 4-4 Determinants of smoking in males: odds ratios for the likelihood of smoking adjusted for all other factors...... 174 Table 4-5 Determinants of smoking in females: odds ratios for the likelihood of smoking adjusted for all other factors...... 176 Table 5-1 The hermeneutic process in analysing company documents ...... 199 Table 9-1 Business practice standards on competition...... 256 Table 9-2 Time line of events...... 260 Table 10-1 Changes in ad-valorem cigarette excise rates over time as a result of BAT’s influence... 278 Table 11-1 Health Decree 30 – the original text and BAT’s response (quoted directly)...... 295 Table 11-2 The original741 and amended decrees 755, 754 ...... 301

8 LIST OF FIGURES Figure 1-1 Trends in LE at birth 1980 to 2000 ...... 33 Figure 1-2 Trends in LE at birth in both genders for regional groupings of countries in the NIS ...... 34 Figure 1-3 The four stage evolution of the tobacco epidemic...... 89 Figure 1-4 Survival by smoking status among men enrolled in the Russian arm of the Lipid Research Clinics study...... 92 Figure 1-5 Probability of dying from tobacco in middle age (age 35-69) in the FSU and the EU, based on 1990 data ...... 96 Figure 1-6 Proportion of all deaths attributed to smoking in men and women aged 35-69,1955-1995 96 Figure 1-7 Incidence of lung cancer per 100,000 1981-2001 for all countries of the FSU...... 103 Figure 1-8 Standardised death rates per 100,000 from ischaemic heart disease, 1981 - 2001 ...... 104 Figure 1-9 Standardised death rates per 100,000 from cancer of the trachea, bronchus and lung, 1981- 2001 ...... 106 Figure 1-10 Trends in the gender gap in life expectancy at birth in the European Union and the Newly Independent States, 1990-2001 ...... 108 Figure 1-11 Rationale for combining methods ...... 118 Figure 3-1 Tobacco leaf production in the USSR/FSU, 1961-2000...... 145 Figure 3-2 Tobacco leaf production in selected countries, 1992-2000 ...... 146 Figure 3-3 Cigarette Production in the USSR/FSU, 1960-2001 ...... 147 Figure 3-4 Cigarette imports in the USSR/FSU, 1960 to 2001...... 149 Figure 3-5 Tobacco leaf imports in the USSR/FSU, 1961- 1999...... 150 Figure 3-6 Cigarette exports in the USSR/FSU, 1960-2001...... 151 Figure 3-7 Cigarette consumption per capita in the USSR/FSU (all ages), 1960-2001 ...... 152 Figure 4-1 Prevalence of current and ever smoking by gender...... 168 Figure 4-2 Current male smoking prevalence by age group ...... 169 Figure 4-3 Current female smoking prevalence by age group ...... 169 Figure 4-4 Odds (and 99% confidence interval) of smoking in men having adjusted for age, area of residence, marital status, religion, education, economic situation and level of social support...... 178 Figure 4-5 Odds (and 99% confidence interval) of smoking in women having adjusted for age, area of residence, marital status, religion, education, economic situation and level of social support...... 179

9 LIST OF ABBREVIATIONS AND DEFINITIONS Abbreviations ATK Alma Ata/Almaty Tobacco Kombinat (Factory) in Kazakhstan BAT British American Tobacco BATUKE BAT United Kingdom and Export CDC Centers for Disease Control and Prevention CEC Chief Executive’s Committee CEE Central and Eastern Europe CIS Commonwealth of Independent States, Originally created in December 1991, now includes 12 of the 15 republics that emerged from the FSU -Azerbaijan, Armenia, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Moldova, Russia, Tajikistan, Turkmenistan, Uzbekistan and Ukraine., that is all but the three Baltic states. CNTC China National Tobacco Corporation CSEC Central and South Eastern European Countries DALYs Disability Adjusted Life Years EBRD European Bank of Reconstruction and Development ECOHOST European Centre on Health of Societies in Transition ESPAD European School Survey Project on Alcohol and Other Drugs EU European Union FAO United Nations Food and Agriculture Organisation FCTC Framework Convention on Tobacco Control FDI Foreign Direct Investment FSU Former Soviet Union GATT General Agreement on Trades and Tariffs GDP Gross Domestic Product GKI Uzbek State Privatisation Agency GYTS Global Youth Tobacco Survey HBSC Health Behaviour in School-Aged Children Survey IARC International Agency for Research on Cancer IHD Ischaemic Heart Disease IMF International Monetary Fund JV Joint venture LLH Living Conditions Lifestyles and Health Project LMIC Low and Middle Income Countries LSHTM London School of Hygiene and Tropical Medicine MOU Memorandum of Understanding NBD New Business Development Unit NGO Non-governmental organisation NIS Newly Independent States – all 15 states to emerge from the former Soviet Union OECD Organisation of Economic Cooperation and Development PM or PMI Philip Morris/Philip Morris International

10 PMI Philip Morris International PW Price Waterhouse (now Price Waterhouse Cooper) SFP Samarkand Fermentation Plant SOE State Owned Enterprise SOM State Owned Monopoly TNC Transnational corporation TTC Transnational Tobacco Company or corporation TTF Tashkent Tobacco Factory UFP Urgut Fermentation Plant UK United Kingdom UPP Uzpisheprom, The Food Industry Association/ State Joint Stock concern for the Food Industry, responsible for five industries including tobacco US United States of America USCEA US Cigarette Export Association USDA FAS US Department of Agriculture, Foreign Agricultural Service USSR Union of Soviet Socialist Republics WHO World Health Organization

11 Definitions Papyrossi A type of cigarette popular in the Former Soviet Union often characterised by a long, hollow mouthpiece that can be twisted before smoking. Such cigarettes are filled about one third full with either pure oriental tobacco or a mixture of tobaccos. They are also hand-rolled from granulated tobacco leaves and from midribs of N. rustica Transnational A business conducting its activities across national boundaries Company or with varying degees of coordination and integration and local Corporationa differentiation of strategy and operations depending on market and business conditions. The major tobacco transnationals are Philip Morris, British American Tobacco, Japan Tobacco International and Imperial Tobacco (following its acquision of Reemtsma).However, although as noted in footnote bin the main text, for simplicity, this thesis refers to all major privately owned international tobacco manufacturers as transnational tobacco companies (TTCs). Multnational A multinational business is one conducting international business Company or and operating in several countries but with little coordination of Corporationa activities and subsidiaries across national boundaries such that subsidiaries operating in different countries are allowed a considerable degree of autonomy.

a Source: Stonehouse G, Campbell D hamill J, Purdie T. Global and Transnational Business: Strategy and Management (Second edition) Chichester: Wiley & Sons Ltd., 2004 and Bartlett CA and Ghoshal S. Managing Across Borders: The Transnational Solution. Boston: Harvard Business School Press, 1989.

12 STATEMENT

I can confirm that the work in this thesis is my own.

Signed: Date:

13 ACKNOWLEDGEMENTS I am indebted to Martin McKee, my supervisor for pointing me on the path to discovering the problems of tobacco consumption in the former Soviet Union and the unique resource provided by tobacco industry documents, for giving me a free rein to explore new ideas whilst still providing constant support, guidance and encouragement. Much of the work that appears in this thesis forms part of two major research projects. The tobacco document work forms part of US National Institutes of Health funded project titled “Globalisation, the tobacco industry and policy influence” (grant number 1 R01 CA91021-01). I would therefore like to thank the US National Institutes of health for funding this research and my salary during this period and my colleagues on this grant, notably Jeff Collin and Kelley Lee and the project administrators, Melanie Batty and Nadja Doyle all of whom were tireless in their support. The surveys formed part of a European Community funded project titled “Living conditions, lifestyles and health (in the former Soviet Union.)”, funded under the FP5 horizontal programme “confirming the international role of community research” (INCO2-copernicus; contract no: ICA2-2000- 10031, project no: ICA2-1999-10074). I wish to acknowledge the work of the LLH study teams in the coordination and organisation of data collection across the eight participating countries, in addition to the teams in Austria and the UK. I particularly wish to thank Joceline Pomerleau, with whom I worked closely on this project for her input and methodological support.

I would like to thank Joy Townsend for her advice on tobacco taxation and her co-review of a number of tobacco industry documents on this issue and a large number of colleagues working on tobacco control in the former communist bloc who have been fantastically supportive. I would particularly like to thank Cornel Radu (Romania) and Irina Zatushevski (Moldova) for working with me on the Moldova documents, Konstantin Krasovsky (Ukraine), Andrey Demin (Russia), Chinara Bekbasarova (Kyrgyzstan) and George Bakhturidze (Georgia) for providing background information and data, and Tatiana Andreeva for translating much of this work into Russian so that it can be disseminated in the region.

I would also like to thank Joy de Beyer of the World Bank for sharing her ideas and an unpublished report with me and the Open Society Institute, particularly Roxana Bonnell,

14 for putting tobacco control on the agenda and working with me to help disseminate this work through their channels in the region.

Finally, I would like to thank my husband for his patience, encouragement and support especially during the final few months of writing this thesis.

The funders listed above cannot accept responsibility for any of the information provided or views expressed in this thesis.

15 CHAPTER 1 Introduction, background, aims and objectives

SECTION A INTRODUCTION The Soviet Union collapsed in 1991, precipitating immense political, economic and social changes – a transition to democracy (albeit imperfect), a sudden transfer from a command to a market economy, and the destruction of the Soviet socialist infrastructure on which the population had relied. The enormous and largely negative immediate impact of these changes on health has now been well documented,1,2 but as yet, little thought has been given to their long-term impact on health. Of particular concern in this regard are the overwhelming changes to the region’s tobacco industry, most notably, the rapid and unregulated entry of the multinational and transnational tobacco companies (TTCs)b and the privatisation of the state owned tobacco companies. Not only is the documented negative impact on health of tobacco enormous, killing one in every two of its long-term users, but even before the political transition, the former Soviet Union (FSU) had the highest rate of premature mortality of any part of the European region.3

TTCs can enter markets in two main ways, by virtue of globalisation of production or globalisation of trade.4 Evidence derived from the industry’s entry to Asia in the 1980s, when the American TTCs forced four Asian countries to open their markets to cigarette imports, shows how liberalisation of trade leads to increased cigarette consumption.5 The underlying economic rationale is that increased competition drives up marketing activity and drives down prices, thereby fuelling consumption. The impact of globalisation of production, through liberalisation of inward investment and privatisation of state owned tobacco companies, has not been studied directly but the economic impacts are likely to be similar, although there is the added issue of local investment, which gives the TTCs a greater degree of economic and political influence. 6

b Various terms and definitions have been used to describe international corporations including tobacco manufacturers(see list of definitions). Generally, a multinational business is one conducting international business and operating in several countries but with little coordination of activities across boundaries. A transnational business is a business conducting its activities across national boundaries with varying degees of coordination and integration. In recent years most international tobacco manufacturers have undergone significant change through mergers and acquisitions and global expansion as described further in section 2.3. As a result, some small, largely nationally focused companies have transformed into major multinationals whilst others have moved from multi- to transnational status even during the course of my work. For simplicity therefore, all major privately owned international tobacco manufacturers are referred to in this thesis as transnational tobacco companies (TTCs).

16 Although the TTCs have been establishing global production facilities assiduously over decades, nowhere has the transformation of the tobacco market been so rapid or profound. As the FSU embraced market economics, in many cases under immense pressure from the international financial community, so the doors to this previously closed market opened to the TTCs. The former Soviet Union (FSU) was the third largest cigarette market in the world7 and the TTCs had been greedily eyeing it for some time.8 Its potential importance, alongside other closed markets (and he must have been including the Chinese market in these figures), was aptly summarised by Mike Pavitt, Rothman’s international spokesman in the early 1990s:

“Until recently, perhaps 40% of the world’s smokers were locked behind ideological walls. We’ve been itching to get at them … that’s where our growth will come from.” 9

Over the ensuing decade the transformation of the tobacco market in the FSU was profound: the Soviet industry was dismantled into fifteen national industries, trade was liberalised with cigarette imports (both legal and illegal) increasing massively, many of the emergent national industries were privatised with the TTCs investing heavily and, above all, the presence of the TTCs was felt most profoundly through the introduction of marketing, a previously unknown phenomenon.

This thesis capitalises on the unique social experiment provided by this transformation of the region’s tobacco industry which, despite its scale, remains hitherto unexplored. It examines the issues raised through a combination of data and methodologies: analysis of routine and ad hoc data on investments and consumption; survey data on patterns and determinants of smoking behaviour in the region; and internal tobacco industry documents. As outlined in more detail later, it aims to understand the impact that the FSU’s transition from command to market economy had on the region’s tobacco industry, on cigarette consumption, smoking prevalence and tobacco control and the role the TTCs played in influencing or benefiting from these changes.

17 This work is of importance for a number of reasons:

(1) The nature and scale of the changes to the region’s tobacco industry, including the scale of TTC investments, knowledge of which contributes to an understanding of their degree of policy influence, has not yet been accurately documented.

(2) The impact of tobacco industry privatisation and investment liberalisation on cigarette consumption and tobacco control remains largely unknown, yet these policies continue to be promoted by important international actors. At the time of transition, privatisation of the tobacco industry, whose product is uniquely damaging to health, was not differentiated from the privatisation of any other industry. Indeed the international financial organisations (IFOs) supported, in some cases even pressed for, tobacco industry privatisation in the FSU and elsewhere, despite having no knowledge of and seemingly having given no consideration to its potential impacts.10 Tobacco industry privatisation continues in other parts of the world, in some instances still under IMF pressure,10 providing a pressing need to understand its potential health impacts and how these may be countered.

(3) Country specific data on smoking behaviour, (particularly by age, sex and area), are widely recognised as prerequisites for designing appropriate prevention policies, 11 , 12 , 13 yet most countries of the FSU lack accurate, recent data on smoking behaviour and, other than the Baltic states, none have data in comparative format.

(4) Although it has been alleged that the TTCs have influenced policy in the FSU,14 this has never been formally studied and this thesis represents the first effort to do so. Similarly, although the TTCs previous entry to other regions has been documented anecdotally, with authors attempting to outline the strategies adopted by the TTCs to penetrate new markets in Asia and Latin America, 15, 16 such efforts have never been studied from the industry’s perspective as the industry’s internal records were not available at the time. Thus, this work provides the first industry-based insights on the rationale for and tactics used to penetrate new markets.

18 (5) Finally, until now, the tobacco industry documents have largely been studied in isolation from other materials. This thesis provides a unique attempt to contextualise and triangulate the industry documents with other data sources, thereby overcoming potential weaknesses of document research and enabling the document findings to be placed in a wider context.

THESIS OUTLINE The remainder of the thesis is structured as follows. Section B of this chapter gives the background to the work undertaken and is split into four parts. Part 1 examines the history of the Soviet Union and its collapse, exploring its economic transformation in some detail in order to address the rationale for and evidence on the privatisation of state owned industries. It also gives a brief overview of the health impacts of transition. Part 2 gives the background to the tobacco industry both in the Soviet Union and globally, examining existing evidence on the impacts of TTC expansion to new markets. Part 3 pulls together all existing data on tobacco use and smoking prevalence in the former Soviet Union while Part 4 examines the health impacts of tobacco use. The final section of this chapter, Section C outlines the aims and objectives of this thesis. Given its considerable coverage, Chapter One is by far the longest in the thesis, and is then followed by 11 other, far shorter chapters.

Chapters 2 and 3 set the scene for the subsequent chapters by analysing the changes that have taken place in the region’s tobacco industry and in trade and production of tobacco in recent years. Chapter 4 provides additional core information, reporting on the prevalence and socio-demographic correlates of smoking in eight former Soviet Republics, based on a series of surveys. Chapter 5 describes and critically assesses the methods used to obtain and analyse the internal industry documents that provide the core material for the following chapters. Chapters 6 and 7 trace the methods employed by the TTCs to penetrate the markets of the FSU. Chapters 8-11 look in detail at the methods they used in two countries, Moldova and Uzbekistan. Finally, Chapter 12 draws together the lessons from this process.

19 SECTION B BACKGROUND PART 1: THE FORMER SOVIET UNION 1.1 The history of the Soviet Union17,18,19 The Bolshevik revolution of 1917 led to the creation, on 30th December 1922, of the Union of Soviet Socialist Republics (USSR or Soviet Union). At first occupying the territory of the Russian Empire, the USSR expanded during the Second World War when the three Baltic States, much of pre-war Poland and part of Romania (mostly comprising present day Moldova) were forcibly integrated. With a ring of protective communist regimes established in eastern Europe, the Soviet Union remained as one of the world’s two superpowers until its demise in December 1991.

Although the Soviet Union accomplished much in the immediate post-war period, with a programme of massive industrialisation in the 1940s and 1950s, from the mid 1960s its development began to fall increasingly far behind that in the west. By the mid 1980s, change had become inevitable. On March 11, 1985, Mikhail Gorbachev became General Secretary of the Communist Party of the Soviet Union. His launch of much needed reforms introduced two now-famous terms to the world - glasnost (openness) and perestroika (restructuring). Yet these proved inadequate to save the USSR and his abandonment of the Brezhnev Doctrine, which had effectively prevented any country from leaving the Warsaw pact, precipitated a string of revolutions in eastern Europe during 1989 that culminated in the fall of the Berlin Wall and the collapse of the Communist system.

1.2 The collapse of the Soviet Union Glasnost, by relaxing censorship and increasing political openness, had the unintended effect of re-awakening nationalist and anti-Russian feelings in the Soviet Union’s fourteen other republics. In the 1989 elections to each republic’s regional assemblies, the first democratic elections since 1917, politicians espousing national self-determination swept the board (although in many cases these were the same individuals who had led the local communist parties previously). By 1990 the republics had begun to issue declarations of independence, with Lithuania the first to do so on March 11th, followed shortly thereafter by Georgia.

20 In August 1991, two days before Gorbachev, by then elected as the first executive president of the Soviet Union, was due to sign a new treaty of union with the republics’ leaders to create a federation in which independent states shared a common president, foreign policy and military apparatus, communist party conservatives staged a coup d’état. They attempted to remove Gorbachev and restore the old regime by force but were met with overwhelming popular protest. When the army refused to give its full support to the junta it collapsed.

Immediately following the coup attempt, the two other Baltic States, Latvia and Estonia, declared their independence and later that month, the remaining ten republics followed. Gorbachev was reinstated, but by that stage his position as executive president of the Soviet Union had become obsolete. His rival, Boris Yeltsin, by then elected President of the Russian Republic, had gained the upper hand, not least because the transfer to democracy had destroyed the Soviet Union as a state whilst making Russia’s own statehood possible.20

In early December 1991 Presidents Yeltsin of Russia, Kravchuk of Ukraine and Shuskevich of Belarus signed a treaty formally disbanding the Soviet Union and creating a new Commonwealth of Independent States (CIS). The Central Asian republics were furious that they had not been consulted yet still agreed to join.21 On 21 December 1991 in Alma Ata, eleven of the former Soviet republics signed an accord forming an economic and political alliance, the CIS, effectively ending the USSR. The three Baltic States refused to join as did Georgia, although the latter then rescinded. On December 25, Gorbachev resigned, on the 26th the Supreme Soviet officially dissolved the USSR and on midnight December 31, the Soviet flag above the Kremlin was replaced by the Russian tricolour.

1.3 The fifteen Newly Independent States (NIS) The fifteen countries to emerge from the FSU differ in many respects – in terms of size, ethnicity, urban-rural balance, economic development, democratic institutions and health status (Table 1-1). Geographically the NIS can be divided into three regions: the European (Estonia, Latvia and Lithuania, Russia, Ukraine, Belarus and Moldova), the Caucasian (Armenia, Azerbaijan, Georgia) and central Asian states (Kyrgyzstan,

21 Kazakhstan, Tajikistan, Turkmenistan and Uzbekistan). The European countries can be further divided into the Baltic States (Estonia, Latvia and Lithuania) and the rest. The Baltic States only became part of the USSR after the Second World War and therefore had prior experience of independence. They have fared far better post-transition than the other former Soviet republics and in May 2004 joined the European Union (EU).

The European countries are more industrialised and developed, particularly compared to the central Asian states, with the exception of Kazakhstan. This is reflected in demographic and health statistics (Table 1-1) which indicate that the European populations have completed the demographic and epidemiological transitions (from natural to controlled fertility and exogenous to endogenous causes of death).22 Thus the central Asian states have high child and moderate adult mortality (a pattern common to middle income countries) whilst the European countries (almost uniquely) have high adult and moderate child mortality, and the Caucasian countries, like others in southern Europe, have moderate mortality rates in both age groups.23 The few exceptions are Kazakhstan which fits the European rather than the central Asian profile and Azerbaijan which more closely fits the central Asian profile. Similarly, non-communicable diseases predominate in the European region whilst central Asia faces a double burden with, in addition, relatively high rates of some communicable diseases.23

Each country’s response to and experience of transition also differed. War and civil strife afflicted Armenia, Azerbaijan and Tajikistan in 1992-4 and Georgia and Moldova in 1992. Hostilities in Chechnya (Russia) continue. The degree of true democracy, with meaningful freedoms, is variable. 24 , 25 The transition to democracy has been most successful in the three Baltic States and least successful in central Asia (particularly Uzbekistan and Turkmenistan) and Belarus.

22 Table 1-1 Selected demographic and economic indicators in the FSU

Date of GNP per joining WTO Mid year % urban % Infant GNP per capita GDP (rest non population population, population mortality capita (US$), growth members as estimate, 1995 1995 0-14, 1995 rate, 1995 (US$), 1995 2001 rate 1994 of Sept 2005) Armenia 3,759,950 69 29 14 730 560 3 05/02/3003 Azerbaijan 7,444,318 56 33 24 480 650 - 22 Observer Belarus 10,280,805 71 22 13 2,070 1,190 - 22 Observer Estonia 1,483,942 73 21 15 2,394 3,810 6 12/11/1999 Georgia 5,368,700 58 22 13 440 620 - 28 14/09/2000 Kazakhstan 16,539,600 60 30 28 1,330 1,360 - 25 Observer Kyrgyzstan 4,481,611 39 38 28 700 280 - 27 20/11/1998 Latvia 2,515,602 73 21 19 2,270 3,260 - 10/02/1999 Lithuania 3,714,795 72 22 12 1,900 3,270 2 31/05/2001 Moldova 4,338,779 52 27 22 920 380 - 22 26/07/2001 Russia 147,773,664 73 21 18 2,240 1,750 - 13 Observer Tajikistan 5,757,480 32 44 31 340 170 - 15 Observer Turkmenistan 4,481,295 45 40 43 920 950 - 16 Ukraine 51,276,560 70 20 15 1,630 720 - 24 Observer Uzbekistan 22,912,100 41 41 26 970 550 - 5 Observer Source: WHO HFA database, IMF website and Bobodilla et al

1.4 Economic transition When the Berlin Wall fell in 1989, one of the world’s most profound economic transitions began, gaining momentum with the collapse of the Soviet Union in 1991. Efforts in the late 1980s to modify and revive the traditional system through partial liberalisation and decentralisation had failed and Soviet leaders had begun to accept the need for more fundamental reform. 26 In July 1990, leaders of the Group of Seven industrial nations and the president of the European Community asked the International Monetary Fund (IMF), the World Bank, the Organisation of Economic Co-operation and Development (OECD), and the newly established European Bank of Reconstruction and Development (EBRD) to study the Soviet economy and propose a method of reform.26 Subsequently published,27 their advice was to pursue a radical programme of rapid and fundamental reform to be implemented in close co-ordination with the IMF: the release of price controls on almost all goods at once, privatisation c of small and, after reorganisation, large enterprises and a lifting of restrictions on foreign trade.26,28 During

c defined as the deliberate sale by a government of state-owned enterprises (SOEs) or assets to private economic agents

23 its final stages the Soviet government recommended reforms, but these were generally less radical than the IMF and others had recommended.26

At the time analysts argued that there was no theory to guide the process of transition.29 The speed of reform constituted a key debate30 and led to the emergence of two schools of thought. The “shock therapy” approach of rapid and extensive privatisation assumed that private ownership by itself would be sufficient to ensure effective reform. Competitive policies and institutional safeguards could follow at a later date. Sachs, drawing on the initial experiences of privatisation in Eastern Europe where rent-seeking, asset-stripping, job-protection and self-appointed wage increases characterised large industrial enterprises that remained in the state sector, suggested that rapid across-the- board privatisation offered the best hope of addressing the systemic crises faced by the state sector.31 Such arguments were also supported by some economists in the region, including those appointed by Yeltsin to spearhead the Russian economic reforms.32 There was fear that delay would might make transition more difficult and thus discredit the potential of a market economy altogether, ultimately leading to a reversion to communism.26,30 The more “gradualist” paradigm argued for a step-wise approach wherein the creation of a competitive environment and necessary institutional infrastructure and regulationd would precede privatisation. China had pursued a gradual transition from the 1970s and achieved impressive successes yet the reformers advising Russia, the US Treasury, and the global financial institutions, particularly the IMF, chose to ignore its experience as well as the advice of other Russian economists and instead preached the textbook economics of market fundamentalism, advocating “shock therapy”30 in what became known as the “Washington consensus”.33

These policies, emphasising rapid privatisation, were based largely on the global orthodoxy of the time - the Thatcher and Reagan era of free market ideology. The Thatcher government had coined the term “privatisation” in 1979 and established its goals to: (1) raise revenue for the state, (2) promote economic efficiency, (3) reduce government interference in the economy; (4) promote wider share ownership, (5) provide

d A real and effective banking system, regulatory frameworks to ensure contracts are enforced, commercial disputes can be resolved, bankruptcy procedures followed and competition is maintained, a land market and land registration and so on.

24 the opportunity to introduce competition, and (6) expose State Owned Enterprises to market discipline.34

Although at the time met with great scepticism, the successful privatisation of British Telecom in November 1984 meant that privatisation became firmly established as a mainstream economic policy. The policy was exported globally by the World Bank and IMF “the new missionary institutions”30 and within the space of two decades it had moved from novel economic policy to global orthodoxy.34

Thus at the start of transition, although privatisation was only a fledgling policy and most empirical evidence at the time led to scepticism if not outright criticism of its impact,e rapid privatisation was a solution recommended to the new governments of the former communist bloc and in most instances was the path they followed.

1.4.1 The scale and speed of reform and privatisation These recommendations led to privatisation, the scale and speed of which was unprecedented. In less than a decade hundreds of thousands of small-scale firms and approximately 60,000 medium and large scale firms were privatised across the former communist bloc, nearly ten times the number of privatisations seen in the rest of the world in the previous ten years.29 The number of new private companies started was also unparalleled, although higher for central Europe and the Baltic republics than the CIS.35 The extent of privatisation varied considerably. By 1997 at least 70% of Gross Domestic Product (GDP) was in the private sector in the Slovak Republic, Hungary, Czech Republic, Albania, Lithuania and Estonia. Poland, Latvia and Russia, Armenia Kyrgyzstan and Kazakhstan were not far behind at approximately 60% although the first three, like the other countries, had completed much of this privatisation by 1995 whilst in the last three, especially Kazakhstan, much of the privatisation was conducted after 1995. Belarus, Turkmenistan and Uzbekistan have seen far fewer privatisations, in keeping with their generally much more limited political and economic reform, with the other CIS states lying somewhere in between.35

e concluding that other methods of reforming SOEs, such as injecting more competition or providing more consistent oversight, might either have been more effective or entail fewer social costs

25 The World Bank has developed a liberalisation index to measure the degree of policy reform, encompassing both privatisation and broader reforms (elimination of central planning, liberalising trade, pursuit of pro-competition policies). Within the CIS the spectrum from most to least reform after ten years of transition has been described as follows: Kyrgyzstan, Kazakhstan, Georgia, Moldova, Armenia, Russia, Ukraine, Azerbaijan, Uzbekistan, Tajikistan, Belarus and Turkmenistan.36

Privatisation inevitably led to a flood of foreign direct investment (FDI) particularly to central Europe and the Baltic states and within the CIS mainly to the energy rich countries, (Russia, Kazakhstan and Azerbaijan) although not Turkmenistan, where President Nyazov was developing a system of government built around an increasingly bizarre personality cult.36

1.4.2 Outcomes The outcomes of this policy were mixed. All transition economies experienced an initial economic downturn (which had not been predicted by the experts) but more recently recovery has occurred and growth resumed in central and eastern Europe (CEE) and the Baltic republics.33 Indeed these countries, now fledgling EU members, have achieved macro-economic stabilisation although by the turn of the century only a few (Poland, Hungary, Slovenia and Slovakia) had achieved a GDP per capita equal to that of decade prior.30, 33 The consequences for the FSU were far less positive. Stabilisation has not occurred, prospects for growth are less certain and market institutions do not function.33 Stiglitz has argued that the consequences for the FSU were little short of disaster.30 In the first 10 years economies declined by up to 60% and by the turn of the decade the CIS as a whole had recovered only 63% of its initial GDP. Individual country declines in GDP were startling – by 2000, GDP in Russia was less than two-thirds of what it was a decade before, in Ukraine one third and in Moldova under a third. The middle class has been decimated and corruption increased. These changes have had enormous social impacts – standards of living fell and poverty increased as the negative impacts fell disproportionately on the poor. Between 1989 and 1998 the proportion of the Russian population living in poverty, using the $2 a day standard, increased from 2% to 24% and comparable changes were seen elsewhere in the FSU.30 Inequalities have increased (doubling in Armenia, Russia, Tajikistan and Ukraine, as measured using Gini

26 coefficients36) reaching levels seen in Latin America, well above those elsewhere in Europe.30

The poor Russian results deserve more detailed study, particularly as similar problems were experienced in many CIS countries. A particularly harsh, but many would argue, accurate critique comes from Stiglitz, whose position as chief economist of the World Bank from 1997 to 2000 gives him unique insights. He suggests that most prices were freed almost overnight in 1992 f leading to hyper-inflation that wiped out savings. Attempts to control inflation led in turn to excessively high interest rates.30 High interest rates encouraged asset stripping, which, facilitated by the absence of systems of corporate governance, and capital market liberalisation (another IMF policy) facilitated the exit of improperly acquired money. The Russian government became increasingly impoverished - not only had it sold its assets for a pittance under pressure from the IMF to act quickly but no effective taxation system was in place - enabling the newly created class of oligarchs and businessmen to pay only a fraction of what they owed in taxes as they engaged in a massive exodus of capital.30,32

In 1998 the situation deteriorated further as the economy was hit by the fallout of the East Asia financial crisis and a fall in oil prices. Despite the disastrous consequences, the IMF insisted that Russia maintain its overvalued currency, fearful that devaluation would set off a round of inflation, until the economy finally crashed. Despite an enormous IMF loan (which within days was showing up in the oligarchs’ Cypriot and Swiss bank accounts), the government devalued in August 1998. According to Stiglitz, the devaluation led to the first significant economic growth, an indication of the failure of IMF policies.30

Stiglitz is similarly critical of IMF influence elsewhere in the former communist bloc, arguing that countries such as Poland that ignored IMF pressure to control inflation, and instead ran it at around 20%, have seen the greatest success. Similarly China, which followed its own, not the IMF’s prescription has seen great success; in 1990 China’s GDP was 60% that of Russia, a decade later these figures had reversed. By contrast the Czech Republic bowed to IMF pressure and used high interest rates to push down inflation, which in turn stifled investment and the economy in general. 30

f A few prices, including those for natural resources, were kept low. Thus, if you could buy oil and sell it in the west you could make millions overnight (as some people did).

27 1.4.3 A retrospective analysis – gathering the evidence on privatisation The concept of privatisation was spreading widely in the late 1980s and early 1990s, with a growing body of evidence suggesting it could bring benefits in some circumstances. However, its failures in Russia in particular and the CIS in general, along with other concerns about its unintended consequences have led economists to reconsider its role as a policy instrument.37

The positive evidence about privatisation shows that private firms outperform state owned enterprises in efficiency and profitability and that privatisation of state owned enterprises leads, in almost all instances to significant and often dramatic improvements in operating and financial performance of divested firms.34,37 While this evidence comes mostly from OECD countries, some multi-country surveys and individual country case studies from low and middle income countries reach similar conclusions.37,38

While this evidence seems overwhelming, some of its limitations must be considered. Firstly, there is scope for selection bias as the best performing firms are more likely to be privatised. Secondly, privatisation may appear to lead to efficiency improvement because it occurs contemporaneously with deregulation or competition enhancement, which some suggest are as, if not more important than changing ownership. It seems intuitive that the context within which privatisation takes place, in particular the economic policy environment and the presence of functioning of legal and administrative institutions that create and enforce property rights and regulate capital markets, must also matter.37 Tandon suggests this is the case and that in many instances where privatisation has not been combined with such changes it does not lead to efficiency improvement, leading him to conclude that it is the level of competition, not ownership, that best determines outcomes.37 The limited empirical evidence that exists in this area suggests that such a combination of circumstances and reform measures can improve the efficiency of State Owned Enterprises.34

Most of this evidence is however at the level of the enterprise. There is little evidence on the role that privatisation plays in influencing economic growth at a national level, - studies testing for determinants of growth do not, for example, include private sector GDP as a variable. And although one study found that countries with sustained growth

28 tended to have a greater share of private sector activity, there were many exceptions to this pattern.35 It must also be noted that few studies have considered the wider or societal impacts of privatisation including its impact on workers, women, the environment and the distribution of wealth and health.37, 39 Some studies document increases and others declines in employment.37 Concerns have been voiced that privatisation might increase inequalities although insufficient evidence precludes firm conclusions on this.37 Other work, both without and within CEE, suggests that women, who already face a large burden of ill health,40 may suffer particularly badly.37

1.4.4 Further evidence from CEE/FSU Consistent with the general evidence described above, a 1998 IMF review of the literature on privatisation in transition economies concludes that privatisation works in as far as privately owned firms generally outperform those that are state run.35 There is a hierarchy of efficiency with start up firms the most successful, followed by outsider dominated firms (especially those with foreign investors) and then insider-dominated firms which outclass those that remain state owned.35 The method of privatisation, which was often dictated by what was politically expedient rather than most economically viable, also matters for another reason – insider-dominated privatisation may generate rent-seeking, oligopolistic vested interests that distort the establishment of an open competitive market and a level playing field. It will, when combined with government privileges such as tax exemptions, lead to a distorted allocation of resources towards the least efficient but most politically favoured.29,35

But privatisation alone is inadequate. The experience from transition countries substantiates findings from elsewhere, that the environment within which privatisation takes place is important. Although some might argue that the environment is more important, current evidence precludes such a conclusion.29 Nevertheless an enabling market environment is clearly vital, with four elements thought to be of particular importance: macroeconomic stability, hard budget constraints (the removal of government subsidies and tax concessions used to prop up inefficient firms), competitive markets and effective property rights.29,35

29 In terms of the pace of reform, it has been suggested that there is insufficient evidence to reach firm conclusions with successes and failures seen amongst both the rapid and slow reformers. 35 Poland, one of the most successful countries to emerge from the transition, is cited in support of a slow approach. It implemented major market reforms in January 1990 (deregulating prices, introducing foreign competition to many industries, and signalling that tight monetary and fiscal policies would be pursued) but delayed its large- scale privatisation programme.41 Indeed, Grzegorz Kolodko an economist and Polish Finance Minister from 1994 to 1997 is one of the fiercest critics of rapid reform arguing that the sequencing of policy measures in much of the region was wrong – privatisation should have occurred only after adequate institution building.33,42 He suggests that the experts advising “shock therapy” lacked experience of and failed to consider the communist legacies and that this led to a failure to give the institutional arrangements adequate import and to an incorrect sequencing of policy measures.33 The experiences of Russia and the Czech Republic, rapid reformers with poor outcomes, the first dramatically so, also support this position. But Estonia and Latvia, by contrast, made rapid and successful progress.35

Finally, consistent with the general evidence described above, the societal impacts of privatisation in transition economies have barely been considered although in the FSU it is apparent that privatisation has generally benefited few at the expense of the very many.37

1.4.5 Success or failure? If conclusions are possible, they are as follows. Privatisation can, and does in many instances, lead to efficiency gains but it has to be done appropriately.35 Thus, although success from privatisation was seen in CEE and the Baltic republics, studies from the CIS have shown few benefits.37 Indeed, the benefits of privatisation decline the further one travels east, in line with evidence that the lower a country’s income the more likely it is that privatisation will go wrong. The problems are particularly acute in the CIS where it appears that failure resulted both from the methods of privatisationg used, the absence of

g good assets tended to go to the well-connected via a variety of dubious schemes (“spontaneous privatisation” that preceded official schemes, manipulation of voucher schemes and via secondary trading). In many instances the well-connected then stripped the assets or sold the licenses as this was a far simpler

30 an appropriate market environment or institutional infrastructure, and the presence of corruption. Failure was more likely where the government was weak or corrupt (or captured by groups who used the state to hide their practices), where there were weak institutional structures or safeguards and little adherence to the rule of law. Whilst the CIS governments then must take their share of the blame, John Nellis of the World Bank also suggests that the international financial institutions must bear some responsibility since they “requested and required transition governments to privatise rapidly and extensively.” Kolodko concurs, stating that transition can only be executed in a gradual manner since building new institutions, laws and changing behaviour, essential to the success of transition take time.42 He includes in his arguments the need to develop a social security safety net, unnecessary under socialism when, for example, formal unemployment did not exist.42 This argument is missing from most other economic reviews, which simply fail to consider the broader societal impacts of privatisation as outlined above.

1.5 Health impacts of transition These major economic, political and social changes following the collapse of the Soviet Union in 1991 have had enormous implications for health.3 Life expectancy, which had stagnated or declined since the 1960s, albeit with a short term increase in the 1980s as a result of Gorbachev’s anti-alcohol campaign,23 assumed a more rapid decline in the early 1990s (Figure 1-1). 43 This decline continued until 1994, was followed by brief but considerable improvement until 1998, but the decline then recommenced, with the changes in life expectancy broadly mirroring the economic crises experienced. The decline in life expectancy was most marked in the European countries and Kazakhstan, and less marked in the other central Asian states and the Caucasus (Figure 1-2).h The exception to the general European pattern was Belarus, which rejected the economic and political reforms undertaken elsewhere, instead retaining a Soviet system of government and, rather than showing a sudden acceleration in the decline in life expectancy and later way of making a fast return. Some valuable assets were sold or given to ordinary workers (e.g. via voucher schemes) who were then pressurised to sell at low prices.

h These graphs also illustrate the lower life expectancy in central Asia and higher life expectancy in the Caucasus.

31 improvement, has followed a steady downward course.3 The sudden fall in life expectancy in Armenia in 1988 was due to an earthquake and in Tajikistan in 1993 due to the civil war (Figure 1-2). Unlike the changes in life expectancy seen amongst other populations in the 20th century, the variations in the FSU have been due largely to changes in adult (and particularly male), rather than child mortality. The decline in life expectancy in the early 1990s predominantly affected young and middle aged men, further widening both the mortality gender gap within the FSU (already the widest recorded anywhere in the world)23 and the gap in life expectancy between the CIS and the EU.44 By 1999 the latter had reached 12.9 years in men and 8.3 years in women. Moreover, in 1993, adult males in Russia, Ukraine and the Baltic republics experienced mortality rates comparable to those of India in 1990 or Guatemala in 1960.23

32

Figure 1-1 Trends in LE at birth 1980 to 2000

80 85 women men EU EU 75 80

70 CSEC CSEC 75

65 CIS

CIS 70 Life expectancyLife at birth (years) expectancyLife (years) at birth 60

55 65 1980 1985 1990 1995 2000 1980 1985 1990 1995 2000 Year Year

Source: WHO Health For All Database updated June 1993. http://hfadb.who.dk/hfa/ (last accessed 8/12/03). Adapted from Nolte, McKee, Gilmore 200443. Code: CSEC – Central and South Eastern European Countries including the Baltics, CIS – Commonwealth of Independent States, EU – European Union

33 Figure 1-2 Trends in LE at birth in both genders for regional groupings of countries in the NIS The Caucasian countries

78

76

74

72

70 Armenia Azerbaijan 68 Georgia

66

64

62

60

1 6 7 9 0 3 5 6 9 8 8 8 9 9 9 9 9 9 984 9 988 9 9 9 9 9 997 9 1 1982 1983 1 1985 198 1 1 1 1 1991 1992 1 1994 1 1 1 1998 1 2000 2001

The central Asian countries (excluding Kazakhstan)

78

76

74

72

Kyrgyzstan 70 Tajikistan Turkmenistan 68 Uzbekistan

66

64

62

60

2 6 7 9 0 2 3 7 0 8 85 8 8 8 9 9 9 95 96 9 0 9 9 9 9 9 9 9 9 0 1981 1 1983 1984 1 19 19 1988 1 1 1991 1 1 1994 1 1 19 1998 1999 2 2001

The European countries plus Kazakhstan

78

76

74

72 Estonia Latvia Lithuania 70 Belarus Republic of Moldova 68 Russian Federation Ukraine 66 Kazakhstan

64

62

60

1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source: WHO Health For All Database updated June 1993. http://hfadb.who.dk/hfa/ (last accessed 8/12/03) 34 Given its large size and the magnitude of its decline in life expectancy, much of the research into the health impacts of transition has focused on Russia. Such work reveals that both the decline in life expectancy in the early 1990s and the subsequent improvement until 1998 were driven largely by changes in cardiovascular and external causes of death (injuries, poisoning, including accidental poisoning by alcohol, murder and suicide), which are unusually high in the FSU compared with elsewhere in the world. 45 It was estimated that 40% of the decline in women and 34% in men was attributable to cardiovascular mortality.46 The groups affected most have been those in lower social classes in regions undergoing the most rapid economic transition. 2,47,48,49,50

Whilst the immediate causes of death responsible for this decline are clearly apparent, the underlying mechanisms remain poorly explained.48 The unprecedented political, social and economic changes experienced with transition have clearly played a role. Material factors such as impoverishment and psychosocial factors, including instability and the stress of change, all contributed, although the precise inter-relationships remain incompletely understood.51 One plausible model envisages a population with an already high level of mortality, in which many deaths were attributable to causes related directly or indirectly to alcohol,1,49,50 causes where the interval between changes in exposure and death were short. In such a population, where major social change rapidly led to changes in the pattern of hazardous drinking, mortality can fluctuate rapidly. This model also envisages a population in which a large proportion have been rendered vulnerable to the effects of rapid transition, 52,53,54,55,56 whether by virtue of their lack of transferable skills, poor social support,2,49, 57 , 58 , 59 or lack of many of the institutional frameworks that facilitate populations elsewhere to adapt to change, such as access to affordable credit and property rights, set against a background, in many countries, of extensive corruption.

Although most work has centred on Russia, more recent work in Ukraine confirms that control over life and material deprivation are important and independent health determinants. It also suggests that change in itself may be damaging to health.60 Control was found to account for the negative impact of low social position on health but not of material position or ‘change’. By identifying these factors as determinants of health, the findings suggest that a decrease in perceived control, arising from an increasingly uncertain political and economic environment, a reduction in material wealth and the

35 stress of change may all have contributed to the decrease in life expectancy seen with transition.

That the profound changes of transition should have had an impact on lifestyles and health should not be a surprise. If little thought had been given to safeguards against market failure, even less had been given to the social safeguards needed in transition. Thus, although in the Soviet Union workers had a “job for life”, which brought with it housing and retirement benefits, no housing market or social safety nets were created in the NIS.30 The very fabric of society on which the Soviet system relied was lost almost overnight.

Although, as we shall explore below, the TTCs, with their highly sophisticated marketing techniques, entered the region in the early 1990s, there is little evidence that smoking can account for the rapid changes in mortality. Smoking behaviour tends to change slowly, and consumption of tobacco, unlike alcohol, which was widely produced illegally, was constrained by severe cigarette shortages in the early 1990s. Most importantly, however, the health impacts of smoking are seen at a population level decades after the onset of smoking and thus any short-term changes in smoking behaviour are unlikely to impact on health statistics in the short-term.

36 PART 2: THE TOBACCO INDUSTRY 2.1 The Soviet Tobacco industry The first reliable account of tobacco smoking in Russia dates from the middle of the 17th century when tobacco was apparently treated as “devilish poison” and numerous imperial decrees ruthlessly punished those who were involved in spreading or using it.61 The first tobacco workshops in Saint Petersburg were opened at the beginning of the 19th century. Papyrossi were first mentioned in an 1844 Russian Finance Ministry circular.61 From the mid 1800s the number of cigarette factories gradually increased and many of the existing production facilities and brand names date from this time. 62 , 63 In 1886 the Russian playwright, Checkov penned a play titled “On the Harmful Effects of Tobacco”. 64 Interestingly, although descriptions of the tobacco epidemic based on trends in the west (largely the UK and US) outline how men in the west began smoking around the turn of the 20th century, historians suggest that the English picked up a cigarette habit during the Crimean war (1854-6),65 in part from their Russian enemies and that many of the hand- rollers in Britain in the 1860s and 1870s were Russian (personal correspondence Matthew Hilton, 16th December 2003).

After the 1917 revolution, tobacco factories were nationalised and the industry became a state owned monopoly. A state body, Rostabakprom, managed tobacco industry enterprises, distributed state investments, directed scientific research and organised centralised imports. From the 1970s onwards the TTCs had a presence in the region, albeit a minor one. Most active was Philip Morris, which had worked with Soviet officials on a number of issues from the mid-1970s7 and whose American-blend brands Apollo Soyuz and Marlboro, were made under licence.66 RJ Reynolds, British American Tobacco (BAT), Reemtsma and Austria Tabak had a smaller presence through imports, but in total the TTCs held under 3% of the Soviet market.7

Branding, marketing and advertising were unknown concepts in the Soviet era. Cigarette brands were not owned but produced jointly across a number of factories. Historically self sufficient in tobacco products, the market was made up largely of traditional filterless cigarettes (papyrossi or oval cigarettes). The production of filter cigarettes did not begin until the 1960s.67,68 It then increased gradually, so that filtered products contributed 21% of total market by 1982 (compared with 94% and 93% in the UK and US respectively in

37 the same year),68 while the proportion contributed by papyrossi declined over the same period from 76% to 30% (the difference being unfiltered cigarettes, Table 1-2).68 Production patterns however varied between republics (Table 1-3).

Table 1-2 Trends in production of smoking materials in the USSR All smoking Papyrossi Cigarettes Filter cigarettes materials (papyrossi (including filter and + cigarettes) non-filter) Year No. (million of No. % of No. % of No. % of % of pieces) (million total (million total (million total cigarettes of pieces) of pieces) of pieces) 1963 257,800 196,000 76.0% 61,800 24.0% 700 0.3% 1.1% 1965 304,400 212,700 69.9% 91,700 30.1% 4,100 1.3% 4.5% 1970 322,700 174,700 54.1% 148,000 45.9% 32,600 10.1% 22.0% 1975 364,300 162,400 44.6% 201,900 55.4% 58,600 16.1% 29.0% 1980 364,000 116,200 31.9% 247,800 68.1% 75,900 20.9% 30.6% 1982 359,400 106,200 29.5% 253,200 70.5% 76,700 21.3% 30.3% Source: Zaridze D. Dvoirin VV, Kobljakov VA, Pisklov VP. Smoking patterns in the USSR. In: Zaridze DG, Peto R (eds). Tobacco: A major international health hazard. IARC Scientific publications No. 74. Lyon: IARC, 1986. 68

Table 1-3 Production of cigarettes as a percent of smoking material production in 1982 for selected Soviet Republics Republic % of total smoking material production contributed by cigarettes Armenia 78 Azerbaijan 99 Estonia 100 Georgia 91 Kazakhstan 54 Lithuania 100 Russia 57 Tajikistan 100 USSR 70 Source: Zaridze D. Dvoirin VV, Kobljakov VA, Pisklov VP. Smoking patterns in the USSR. In: Zaridze DG, Peto R (eds). Tobacco: A major international health hazard. IARC Scientific publications No. 74. Lyon: IARC, 1986. 68

Table 1-4 Tar and nicotine yields in USSR, US and UK cigarettes in 1986 Tar Nicotine Mg/cig Median Mg/cig Median US 1-27 14 0.1-1.8 1.1 UK 17-28 21 0.9-1.8 1.1 USSR 21-31 25 1.3-1.9 1.6 Source: IARC 69 (USSR data from the Laboratory of the Government Chemist)

38 Tar and nicotine yields were considerably higher than those in the UK and elsewhere. An assessment of 41 cigarette and papyrossi brands sold on the USSR market in the 1980s found that 90% of brands had “very high” yields (20mg/cigarette and over) as defined by IARC.68,69

By the 1980s the USSR had adopted a wide range of anti-smoking policies. Advertising was banned entirely and smoking in many public places including subways, buses and restaurants was forbidden.70 Cigarette packages carried health warnings70 although they omitted explicit information about the diseases caused by tobacco.68 Although there is much uncertainty about the effectiveness of Soviet health promotion campaigns (see Section 4.8), anti-smoking campaigns were run on television. There was, however, no state tobacco control programme nor any medical assistance for smokers to quit71 and cigarettes were extremely cheap.

2.2 Transition and entry of the tobacco transnationals

The 1960s had seen a gradual increase in cigarette imports, largely from Bulgaria, and the USSR became the largest importer of cigarettes in the world.7 By the mid 1980s cigarette production had fallen but economic difficulties prevented an increase in imports, leading to an under-supply of cigarettes. 7

Reasons for these shortfalls have been debated. It is clear that lack of investment, outdated machinery and shortage of inputs (whether seed, leaf, paper or filters) played a key role.7 What is uncertain, however, is whether this was simply the result of economic hardship or whether, as part of his health campaign at the time, widely believed to focus only on alcohol, Gorbachev had also deliberately aimed to reduce tobacco output. It seems more likely he used the latter to hide the extent of the economic downturn. Detailed analyses of the anti-alcohol campaign do not mention any action on tobacco;72 and the justification for the anti-alcohol campaign, the major losses of productivity through drunkenness, was quite specific.

By the turn of the decade the tobacco industry was in disarray. Approximately half of the USSR’s cigarette factories were closed and cigarettes were in very short supply. 73 Smokers in Moscow, Leningrad, Kiev and other Soviet cities queued through the night but still came away empty handed, finally staging protests that became known as the

39 “tobacco riots” or “rebellion”. Gorbachev pleaded with the west for help and Philip Morris and RJ Reynolds came to the rescue, keen to get a foothold in this market. Between 1990 and 1991, thirty-four billion cigarettes were airlifted to the FSU, the single largest export order in the history of the tobacco giants.74 In the first 6 months of 1991, it is reported that cigarette exports to the FSU from the US alone increased more than 7200%.75 Thus cigarette imports, which had fallen in the late 1980s, saw a sudden surge of both legal and illegal products, many of dubious quality and lacking health warnings.73,76,77

Having secured the import of its brands, the industry then moved to acquire failing state- owned factories and by the mid 1990s local production began to rise once again.78,79,80 Unlike the entry of the industry to Asia, there was little government opposition to the transnational companies, who were welcomed as creators of economic well-being while public health concerns were shunted aside.81

Despite the scale of the change to the FSU’s tobacco industry, there has been no systematic research into its impact. Nevertheless, reports of success in tobacco industry journals, anecdotal reports of concern in tobacco control journals and reports from the US Department of Agriculture (USDA) combine to outline the nature of many of these changes.

As part of the overall increase in cigarette production, 82 filter production increased disproportionately and now accounts for the vast majority of production in most of the NIS (Table 1-5). Papyrossi production has meanwhile declined, falling to just 5% of the total in Russia in 1999. 83 As a result of these changes, (and in some countries the introduction of new standards on cigarette yields), tar and nicotine levels in most of the region have fallen.

40 Table 1-5 Production of filter cigarettes as a percent of total cigarette production in 1996-2001 2001 AVERAGE 1996 1997 1998 1999 2000 (est) 1996-2001 Armenia 100% 100% 100% 100% 100% 100% 100% Estonia 100% 100% 100% 100% 100% 100% 100% Georgia 100% 100% 100% 100% 100% 100% 100% Kazakhstan 100% 97% 97% 97% 98% 97% 98% Kyrgyzstan 30% 30% 31% 32% 32% 32% 31% Latvia 100% 100% 100% 100% 100% 100% 100% Lithuania 100% 100% 100% 100% 100% 100% 100% Moldova 25% 21% 21% 33% 29% 29% 26% Russian Federation 71% 67% 60% 61% 64% 67% 65% Uzbekistan 3% 41% 45% 55% 60% 60% 44% Source: USDA

The introduction of new brands, notably new middle price and expensive brands targeted at the emerging middle class, required marketing back-up and within a short period of time tobacco advertising became ubiquitous, often flaunting existing tobacco control legislation. 14,84 The industry journals themselves described the streets of Moscow as “a battle ground in a cigarette war of words”.63 Authors report that in 1993 40% of all foreign advertising in Russia was for tobacco, that foreign cigarette brands became the leading advertisers on Russian television and radio74 and that such advertising particularly targeted children and young people, using television adverts screened in the evening.73 In addition to the usual billboard advertising, Philip Morris has been giving away ‘desirable goods’ and BAT used dancing girls to distribute samples outside metro stations. 63

There are grave concerns that in a region virtually unexposed to western type advertising, the population may be more sensitive to the novelty and glamour of tobacco advertising.74 As Vitaliy Movchanyuk, director of the Ukrainian Health Ministry’s public education institute said

“The Soviet Union never had such advertising. People are used to it in the West. They have learnt to sift through it for truth and lies… But our consumers are psychologically vulnerable to being manipulated by slick advertising.” 74

The industry is clearly aware of this; industry journal reports note that smokers “are vulnerable to cigarette advertising”, even reporting that, as a result of successful

41 advertising campaigns by the main tobacco players, “Russians see smoking as the distinction between human beings and animals.” 63 Regional experts have suggested that faced with such “sophisticated and ruthless promotion”, it became increasingly difficult to control the use of tobacco.85

The industry was initially quick to seize on the population’s desire to “westernise” promoting their products as an indispensable part of the “western lifestyle”. As early as 1989, Reemtsma organised a campaign entitled “West goes East” in Moscow’s Red Square to promote their West brand.7 Others used slogans such as “A date with America” for L&M and “Go for It” for Hollywood.86

2.3 The history of the international tobacco industry and its global expansion The TTC’s entry to the former communist bloc is just the latest geographical expansion of an industry whose globalisation began with the formation of BAT in 1903 and escalated from the 1950s onwards.

Although the history of tobacco use stretches back to the first century AD amongst the Mayan people of Central America, reaching Europe via America and the Caribbean islands at the end of the 15th century,87 it was not until the early 1880s that tobacco use really escalated.65 The stimulus for change was the introduction of the Bonsack machine. Mechanisation and the mass manufacture of cigarettes stimulated cigarette marketing, initially to overcome resistance to machine-made goods and then to expand demand in order to accommodate the vastly increased production.15 Competition between the companies in each of the US and UK markets escalated, prompting the first series of tobacco company mergers. The five leading US cigarette manufacturers merged to form the American Tobacco Company (ATC) in 189015,88 and in 1901 the major British firms joined to form the Imperial Tobacco Company (ITC).89 In 1903 ATC and ITC agreed to keep out of each others territories and combined their overseas operations in a new London-based company, British American Tobacco Ltd (BAT), which took over all the business outside the UK and US (plus Cuba and Puerto Rica),15 thus enjoying a virtual

42 monopoly of the international markets.i BAT has thus from its outset been a global company.

By contrast, the first ventures abroad by US-based firms occurred from the 1950s onwards. The stimulus at this time is widely believed to have been the initial health scares that accelerated with the 1964 Surgeon General’s Report on Smoking and Health which caused considerable alarm and a decline in sales. 15,90,91 More recent expansion has been fuelled by global political and economic change including trade and investment liberalisation and the opening of formerly closed markets. From the mid-1960s and throughout the 1970s, the focus was on Latin America. Western Europe was also a target in the 1970s.92 In the 1980s the focus shifted to Asia and in the 1990s to the former eastern bloc as this significant market opened. China, the world’s largest market, is the latest target with companies jostling to establish the first substantial joint venture with China National Tobacco Corporation (CNTC).93

The pattern of mergers and acquisitions established early in the last century has intensified over time, with 141 mergers or acquisitions by major tobacco multinationals since 1990.6j Driven by the need to harness potential economies of scale, reach new markets and reduce the threat of litigation, recent examples include BAT’s merger with Rothmans in 1998, Japan Tobacco International’s (JTI) acquisition of RJ Reynold’s international business in 1999, and Imperial’s acquisition of the German manufacturer Reemstma. It has led to a gradual consolidation of the major TTCs to the extent that 75 percent of the world cigarette market is now controlled by just four companies.94 These are Philip Morris, BAT, Japan Tobacco International and the CNTC.94, 95 Excluding CNTC, whose share is explained by its virtual monopoly of the enormous Chinese market, Philip Morris is the market leader, with an estimated 16.5% share of the world market.94,96 BAT is the second largest company and traditionally the most international with over two-thirds of its cigarettes sold in Africa, Asia, Latin America and Eastern

i In 1911 however, the Supreme Court in America ordered the dissolution of ATC into a number of competing firms (including Liggettt & Myers, RJ Reynolds, Lorillard and a reformed ATC) that were then free to compete on the international market. It also led ATC to shed its shareholding in BAT and left BAT as a British-controlled firm. j It is this more than anything which has driven the transformation of smaller tobacco companies into multinationals and the transformation of multinational tobacco companies into transnationals.

43 Europe.97 However, with Philip Morris’ recent worldwide expansion, it is thought that both now sell in over 180 countries (Table 1-6). Table 1-6 The world’s largest tobacco transnationals Cigarette (billions) Cigarette production (% No. of countries in of world total) which active Philip Morris 887.3 16.5 >180 British American Tobacco 807.0 15.0 >180 Japan Tobacco International 447.9 8.1 170 Source: Link and Rossell, Tobacco Journal International 2001, 96

2.4 What can be learnt from the TTC’s previous expansion to new markets? In late 1988 it was estimated that closed markets accounted for approximately half of all the worlds’ cigarettes sales.92 National monopolies have traditionally protected themselves from foreign competition though protective trade measures including bans on foreign imports, high tariffs or import quotas and restrictions on the manufacture, distribution, sale and advertising of foreign brands.15,92 Economic theory predicts that such barriers to trade are likely to reduce the total supply of tobacco products whilst raising the amount supplied by domestic producers and will result in higher prices.5,98 Domestic monopolies are considered inefficient compared with TTCs. They produce cigarettes that use harsher local leaf and tend to be harder to smoke. There is generally no advertising as the lack of competition deems it unnecessary.92 Combined, these trade restrictions and market inefficiencies may therefore have the largely unintended public health benefit of lowering consumption.5,92 Indeed low per capita consumption rates combined with high male but negligible female smoking rates are characteristic of such closed markets.92 Moreover, restrictions on advertising make it harder for new firms to enter these new markets successfully.

According to Shepherd, organisation theory suggests there are three main barriers to entry: the absolute cost advantages of existing firms, economies of scale and consumer preferences for the products of existing producers. He argues that the first two are relatively unimportant to the cigarette industry, whilst the last appears to be vital. Consumer preferences may arise through physical differences in cigarettes or packaging, which tend to be easily copied and thus do not confer long-term advantage, or through the creation of a favourable brand image through marketing. Potential competitors wishing to

44 enter a market have to incur advertising expenditure over and above that of established firms in order to penetrate the market and overcome the stock of “goodwill” or brand loyalty that has accrued to existing brands. This demand creation effort, Shepherd argues, is the main barrier to entry, acting as a deterrent even for large firms that have the necessary demand creation expertise.15 This may explain the phenomenal success of Philip Morris, which has grown from a minor player in the US market in the 1950s to become the world’s largest cigarette company, largely on the back of a single brand, Marlboro.94 Described as “one of the quintessential global brands”,99 its global rise was intimately linked to its successful marketing, with the Marlboro Man declared by Advertising Age to be the number one advertising icon of the twentieth century. 94,100

TTCs have sought to remove these barriers and gain entry to closed markets, either through local production or imports. Local production has certain advantages. It allows the TTCs to avoid import duties and to benefit from lower costs of raw materials, labour and transport, alongside the sometimes less stringent restrictions on tar and nicotine levels. Reduced production costs can in turn enable lower prices, which will promote sales. The TTCs have successfully used both methods; their expansion to Latin America an example of the former and their expansion to Asia of the latter. Each is now considered in turn.

2.4.1 TTC expansion to Latin America In Latin America the focus was on local production. Shepherd suggests that licensing was used in smaller markets and denationalisation and complete take-over in larger markets with greater growth potential. In a licensing arrangement the state firm produces and sells international brands. By legalising sale of the international brand, this reduces contraband whilst preventing the foreign manufacturer from acquiring a direct financial interest thereby offering some advantages to the monopoly holder. But licensing is often used as a forerunner to acquisition – the TTCs establish a licensed production, provide manufacturing technology and agricultural advice and then push the local government to denationalise, leaving the TTC poised to acquire the national company.16,92

A number of tactics are used to push governments to accept these licensing arrangements and thus in turn privatise their tobacco industry. A key tactic is the use of smuggling.

45 Shepherd outlined how smuggling reaches its peak around the time of TTC entry to a market but was unable to implicate the TTCs directly in this process. 15 Smuggling acts as a market softening technique; it creates a demand for the smuggled international brand (often seen as highly desirable) before a domestic manufacturing presence is established, undermines local firms (which can then be more easily acquired) and makes it easier to argue the need for local production on the basis that the demand for quality products has led to the illegal supply. 15,92,101 Shepherd observed that national firms were seriously weakened by the TTCs, unable to compete with their intensive advertising and predatory pricing practices and the TTCs rapidly gained the majority share of the markets in which they established subsidiaries.15,92

2.4.2 TTC expanstion to Asia In contrast to the focus on local production in Latin America, in Asia, the US-based TTCs focused on import penetration. Between 1985 and 1990 the TTCs, assisted by pressure from the US government, successfully pushed four closed markets in Japan, Taiwan, South Korea and Thailand, whose tobacco industries were controlled by government run monopolies, to open to imports. Using Section 301 of the 1974 Trade Act,k the US government, in some instances in co-operation with the US Cigarette Export Association (USCEA),l threatened trade sanctions against these countries alleging that import quotas, high taxes or other restrictions unfairly limited the market to US tobacco products.92,98 In the face of US threats, Japan, Taiwan and South Korea almost immediately removed restrictions on tobacco imports with Taiwan and South Korea also weakening their advertising restrictions. Although advertising in Japan was not restricted by legislation, the US Trade Representative had already forced Japan to abolish its budgetary ratio concept in which the volume of advertising allowed was based on sales in the previous year, thereby disadvantaging US tobacco companies. 102 Thailand put up far greater resistance and its case was resolved differently; the complaint was referred to the General Agreement on Tariffs and Trade (GATT) dispute resolution process with the US Trade Representative arguing that the import barriers and comprehensive advertising restrictions (which the TTCs were constantly breaching) were a violation of the GATT

k strengthened by the Trade and Tariff Act of 1984 l Formed by Philip Morris, RJ Reynolds and Brown & Williamson in 1981 to facilitate their ability to compete more effectively in foreign markets.

46 principles.92,98,103 In late 1990 the GATT Council agreed that the ban on imports violated the trade treaty but upheld the Thai government’s right to use public health policies to protect health as long as they were applied evenly to domestic and foreign products, even where such polices (e.g. advertising bans) would make it more difficult for new foreign firms to compete with existing domestic firms. In late 1990 Thailand lifted its import ban and imports began in 1991. Shortly thereafter additional tobacco control legislation was adopted and tobacco taxes increased in an attempt to further dampen the sales environment. In all four markets the impacts were broadly similar. US cigarette imports increased (overall it is estimated US cigarette exports to Asia increased by 75% in 1988 alone) as did the US cigarette companies market share, the markets became more competitive with marked increases in advertising and cigarette price cuts.91,92,98,103,102,104 In Thailand for example the US TTCs circumvented the fairly comprehensive advertising ban and cut prices in order to absorb tax increases in what has been described as “virtually dumping”.103 In Japan cigarettes moved from fortieth to second most advertised product on television as the TTCs took advantage of the lax advertising controls and the TTCs again absorbed the tax increases so that real prices fell.104 Changes in smoking prevalence - cessation of previous declines and increases in smoking rates particularly amongst women and young people have occurred in all four countries.98,104,105 In South Korea for example smoking rates among male teens rose from 18.4% to 29.8% in a single year, and among female teens they more than quintupled from 1.6% to 8.7%.105 Declines in sales have also pushed the national monopolies to become increasingly competitive, changing their business model in order to survive.103 The most extreme example is the previous Japanese Public Monopoly Corporation which was privatised in 1985, the year the market opened and in 1999 purchased RJRI, the international operations of RJ Reynolds, creating a new company, Japan Tobacco International (JTI) now one of the largest TTCs.104 The interim public company, Japan Tobacco Inc’s response to the increased competition was to introduce seven new brands in 1987, largely targeted at women and adolescents, and to increase the volume and sophistication of its marketing.102

47 2.4.3 Impacts of TTC entry to Latin America and Asia Similarities between the TTC’s entry to Latin America and Asia are notable and reviews by observers in these two regions suggest that once the market is opened a number of key changes take place (Box 1-1). The TTCs attempt to guide production and consumption patterns away from local products to TTC product forms (international filter brands, usually American blend cigarettes) thereby transferring the advantage from local firms to the TTCs and strengthening the already debilitating impact of smuggling to further weaken local firms. 15 Such brands tend to be smoother and easier for new smokers to smoke,105 helped by the judicious use of a complex cocktail of additives including sweeteners and chocolate that may help make cigarettes more palatable to first time users and increase the amount of nicotine delivered, and so the addictive potential, making it easier for them to acquire the habit.106 Intense price and non-price competition usually occurs in which the TTCs, with their ability to withstand price discounting in one market and slick marketing skills, usually triumph.15,92,103 The enormous increase in advertising and promotion, in particular targeted advertising to groups with previously low levels of consumption, most notably women and young people, combined with lower prices, tends to lead to an increase in per capita consumption and a change in smoking patterns, with rates increasing among women and the young. 16,92

These changes are further exacerbated by the TTC’s attempts to undermine national tobacco control policies.15,16 MacKay argues that TTCs are more likely than national firms to deny the health evidence of the impact of tobacco and challenge health initiatives. This may be compounded by the pre-occupation of governments with other health issues, lack of experience in dealing with powerful transnationals and a virtual absence of policies on tobacco. Ironically, the governments that permit these changes often do not even benefit from them (although individuals may through various incentives) as favourable tax regimes are often negotiated and revenues are returned to shareholders in the west.16

48 Box 1-1 Impact of TTC entry on tobacco consumption Experience in Latin America and Asia suggests that the entry of the TTCs and replacement of a small unsophisticated industry with a powerful sophisticated transnational leads to increased consumption through a number of mechanisms91,92: • Price competition • Slick new promotional strategies • Targeted advertising to groups with previously low levels of consumption • Introduction of new smoother brands that are more attractive to smokers and easier for new smokers to smoke • Powerful political lobby against tobacco control measures

2.4.4 Globalisation of trade: economic theory and econometric evidence The largely descriptive evidence outlined above is consistent with economic theory, which predicts that increasing trade liberalisation will likely lead to greater competition in the market and thus reductions in the price of tobacco products, and increased advertising and promotion of these products. The latter results both from the entrant’s efforts to gain a foothold in the newly opened market and increased activity by existing firms attempting to maintain their market share.5 Given the inverse relationship between price and consumption, as well as the positive relationship between advertising/promotion and demand,101 cigarette smoking will likely increase.107 Although the bilateral trade agreements reached between the US and the four Asian countries described above are the best known, a variety of other bilateral, regional and multilateral trade agreements have reduced barriers to trade in tobacco and tobacco products, and contributed to the global expansion of trade of these products, seen largely since the mid-1980s.5,108 For example, after the GATT Uruguay round in 1994, global un- manufactured tobacco exports increased by 12.5% by 1997, after a decade of virtually no growth and global cigarette exports by 42% by 1996. These changes are likely to have contributed to the 5% growth in global cigarette consumption seen during this period.5 Importantly, descriptive evidence and economic theory have been substantiated recently through empirical work including econometric analyses of the impact of trade liberalization. Chaloupka and Laxuthai, the first to examine the issue econometrically, found that in the four Asian economies described above, consumption of cigarettes per person was 10% higher on average in 1991 and the US market shares 600% higher than they would have been if these markets had remained closed.98 Hsieh, Hu and Lin studied the impacts in Taiwan and found that an 811% growth in the market share of imported

49 cigarettes led to a 20% increase in per capita consumption and, alongside increasing overall consumption, had encouraged smokers to switch from domestic to imported cigarettes.108 A subsequent analysis among 42 countries also found that trade ‘openness’ was positively related to cigarette consumption, implying that trade liberalisation leads to increased consumption. Moreover the relationship was strongest in low income countries, of intermediate strength in middle-income countries and non-significant in high income countries implying that trade liberalisation has the greatest impact in low and middle- income countries and no effect in high income countries.5 Similar findings were seen in further work by Perucic and Guindon who examined 80 countries over the period 1970 to 1990 and found that import penetration (the best proxy used for openness to tobacco trade) positively contributed to cigarette consumption in low and middle-income countries. 108 A consensus therefore appears to be emerging that the removal of trade barriers tends to introduce greater competition which, through a lowering of prices and an increase in advertising and promotion, stimulates demand. 109

2.4.5 Globalisation of production As noted at the beginning of this section, TTCs can access new markets through imports or by establishing or acquiring the facilities to produce in-country. It should be noted that whilst global trade has been increasing steadily since the Second World War and markedly so in recent decades as a result of growing trade liberalisation, foreign direct investment (FDI) has also been increasing. In the last decade, FDI has grown considerably faster than trade, not least as a result of investment liberalisation in the former eastern bloc. Yet little if anything is known about its impacts on tobacco consumption. To date only one study has attempted to examine the impact of FDI on tobacco consumption. Using a very indirect measure of investment liberalisation – exchange rate distortions, indicative of a disincentive to invest – it suggested that an increase in exchange rate distortions lead to a decline in cigarette consumption, implying that FDI would lead to higher cigarette consumption.108 The theoretical impact of FDI is of course similar to that of trade liberalisation (the description of changes in Latin America are for example similar to those seen in Asia). However, it may have additional advantages for the investing companies who can benefit from the lower production costs

50 that arise through cheaper labour, inputs and transport costs and the avoidance of import taxes. The increased efficiency and productivity that generally result offer potential for greater profits and increased sales if the lower costs are transferred to consumers. Shepherd also suggests that demand creation cannot be fully exploited through exports as it requires local facilities and marketing and close ties with distributors. Although subsequent experiences in Asia may refute this suggestion, it is possible that the potential for demand creation through local investment may be greater – an existing tobacco company or factory is generally bought with an existing market share, an established distribution system and access to brands with established consumer loyalty. Such brands can of course still be manipulated to make them more attractive to new smokers. But above all, FDI gives the investing company more economic and political leverage within the country concerned than do imports.6 The massive changes seen in the tobacco markets in the FSU, most notably the TTC’s investments and establishment of local production, provide a unique opportunity to examine the impact of FDI on tobacco trade and consumption. The importance of understanding such impacts is underlined by the IMF’s continued pressure for tobacco industry privatisation, for example in Moldova, Thailand, Turkey, Bulgaria, Korea and Mali10 and the fact that a number of other countries retain state- owned monopolies, some of which are currently being considered for privatisation (for example Macedonia, Serbia, Montenegro and possibly Romania) or could in future be privatised (for example Iran, Morocco, China, Taiwan). Indeed it is believed that the remaining monopolies account for 40 percent of the world’s total cigarette consumption.110 The IMF’s support for tobacco industry privatisation, explored further in Chapter 8, is such that it has even made privatisation part of its loan conditions. The World Bank takes a far more progressive stance on tobacco control issues and has been playing a key role in the tobacco control debate, most notably in highlighting the negative economic impacts of tobacco use.109 Since 1991 the Bank has had a clear policy on tobacco (see Box 1-2).109 Yet it too has acted with the IMF in using its loans to pressure for tobacco industry privatisation.10 The Bank has also been central to the current focus

51 on demand side tobacco control measuresm, suggesting that efforts to control the supply of tobacco should be limited to attempts to control smuggling.109 Broadly, it appears that both organisations failed to consider the potential public health impacts of tobacco industry privatisation, embracing it as one element of a universal approach to privatisation. Nowhere in the extensive literature on privatisation is anything published on the costs and benefits of privatising tobacco industries, whose products are uniquely damaging to health. It has been suggested that the IMF’s rationale for privatising state-owned tobacco industries is that it frees governments to pursue more effective tobacco control policies than they would otherwise have done when directly engaged in selling tobacco (a fact confirmed by my own correspondence with Peter Heller of the IMF, October 2005), although there is no empirical evidence to support such a view.107 Of major concern is the fact that even when members of the US Congress wrote to the managing director of the IMF expressing concern about its potentially serious public health consequences, the IMF response was to state that its reading of the research “does not support the contention that privatization of state-owned tobacco companies, per se, is a major cause of increased tobacco consumption”.10 In other words, not only did the IMF fail to consider health issues prior to its promotion of privatisation, even when such issues have been directly raised with the organisation, it has thus far failed to acknowledge them.

m Tobacco control measures can be broadly divided into those that seek to influence the demand for tobacco products (eg controls on advertising) and those that seek to influence the supply of tobacco products. Other than controls on smuggling, the latter are considered more controversial.

52 Box 1-2 The World Bank's Policy on Tobacco

The World Bank has since 1991 had a policy on tobacco, in recognition of its harmful effects on health. The policy contains five main points. First, the Bank's activities in the health sector, such as policy dialogue and lending, discourage the use of tobacco products. Second, the Bank does not lend directly for, invest in, or guarantee in-vestment or loans for, tobacco production, processing, or marketing. However, in a few agrarian countries that are heavily dependent on tobacco as a source of income and of foreign exchange earnings, the Bank aims to deal with the issue by responding most effectively to these countries' development requirements.

The Bank aims to help these countries diversify away from tobacco. Third, the Bank does not lend indirectly to tobacco production activities, to the extent that this is practicable. Fourth, tobacco and its related processing machinery and equipment cannot be included among imports financed under loans. Fifth, tobacco and tobacco-related imports may be exempt from borrowers' agreements with the Bank to liberalize trade and reduce tariffs. The Bank's policy is consistent with the arguments for ending subsidies made in this report. However, the emphasis on supply-side measures has not reduced tobacco consumption in any measurable way from 1991 to today. In the interim, the Bank's work on tobacco control, comprising about 14 countries with total project costs of more than US$100 million, has largely been on health promotion and information. Extending this work to focus on pricing and regulation was sup-ported in principle by the Bank's 1997 Sector Strategy Paper. This report confirms the importance of focusing on price as an effective means of reducing demand. Source: Chapter 7: Curbing the epidemic. Governments and the Economics of Tobacco control109

2.5 The global shift in tobacco related disease patterns Globalisation of the tobacco epidemic is proceeding to such an extent that the decline in tobacco consumption in high-income countries seen since approximately 1970 has been more than balanced by an increase in low- and middle-income countries. This has in turn led to a shift in the burden of disease from high to low income countries.109 Lopez and Murray estimate that, in 1990, 53% of the worldwide burden of disease (measured in Disability Adjusted Life Years, or DALYs) attributable to tobacco was in the developed world; by 2020 this will have fallen to 23% and the burden in the developing world increased correspondingly.111 This worldwide shift has been accompanied within Europe by a shift from western to southern and eastern Europe and the former Soviet Union.

53 PART 3: SMOKING IN THE FSU 3.1 Measuring tobacco use: tobacco consumption and smoking prevalence Tobacco use can be measured using both cigarette consumption and smoking prevalence; the two are inter-related. The disadvantage of consumption data is that it fluctuates with price and incomes and says little about the number of people smoking. Today, smuggling is the major issue limiting the use of consumption data as only legally imported or sold cigarettes are included in consumption figures.

Prevalence data are probably a better marker of smoking behaviour at a population level. They fluctuate less on a short-term basis and are more relevant to policy formulation as they help direct appropriate policies and programmes and assess their effectiveness.112 However, population surveys are required to provide accurate prevalence data.

3.2 Consumption and smoking habits According to Zaridze, per capita consumption in the USSR, based on sales data, reached a maximum between 1976 and 1980 and then stabilised through the 1980s at approximately 1600 sticks per capita.68 In the late 1980s consumption declined due to the economic difficulties outlined above and this decline continued through the early 1990s.67 No reliable published analyses of consumption trends are available beyond that time.

3.3 Smoking Prevalence – adults

There are very few accurate published data on smoking prevalence for the FSU. Routine surveys are rare, being conducted only in the Baltic States and since 2000 in Ukraine as part of a new household survey commissioned by the State Statistical Committee.113 In the Baltic states, the Finbalt Health Monitor project collects smoking data as part of a wider health survey that grew from the long-standing collaboration between Finland and Estonia114 where surveys started in 1990, and later extended to Lithuania (1994) and Latvia (1998).115,116 Other information comes largely from ad hoc surveys undertaken as part of specific research projects.

A growing number of international databases available on the World Wide Web provide smoking prevalence data. These include the World Health Organisation’s (WHO) Health for All database which in 2000 provided sex specific data on only four countries in the

54 region, Belarus, Estonia, Kazakhstan and Russia, although by 2002 these data had been expanded. 117 Other databases include the Tobacco Control Country Profiles (TCCP) compiled by the American Cancer Society,118 the Tobacco Fact File which went on line in 2003, 119 WHO Europe’s Tobacco Control Database 120 and the National Tobacco Information Online System (NATIONS) Database.121 In many instances, however, these databases provide insufficient details to be sure of the provenance or validity of the data and despite the attempts that database designers have made to ensure their comprehensibility, many omit historical data. For example, the data quoted for Ukraine in TCCP are not nationally representative but from an urban sample in Kiev although this is not stated in the database (personal correspondence Konstantin Krasovsky).

Additional data were therefore sought through literature and internet searches and through contact with national experts and WHO staff. Such data sources often have the advantage over databases of providing information on which data quality can be assessed.

Data for the Caucasus are scarce (Table 1-7). The only historical data (i.e. pre 1990) are from a study by Orlovski quoted by Zaridze68 but the data are inconsistent with later surveys finding higher prevalence rates in both sexes, suggesting the samples were in a younger age group. The disparity between the 68.6% male and 4.7% female prevalence rates reported in this 1963 study in Azerbaijan and the far lower rates reported 1999 (which were taken from a household budget survey)118 makes it almost impossible to assess smoking prevalence there. Inconsistencies in data reported from Armenia led WHO to rely on anecdotal reports that smoking is about 50% or greater among males (probably around 60-65%), low among middle-aged women, and starting to increase among young women.122 These estimates have been confirmed in a more recent survey which reports rates of 67.5% in men and 3.1% in women. 123 In Georgia recently published surveys conducted in the capital, Tbilisi suggest prevalence rates of around 55- 60% in men and around 15% in women. 124,125124,125 Similar findings are reported from other surveys.123 In Moldova the prevalence in men has been around 40-45% since 1995 and in women three sets of data in the last few years estimate prevalence at 2-3% suggesting the 18% recording in 1999120 was perhaps a one-off urban sample (Table 1-7).

55 Data for central Asia are even more scarce (Table 1-8). There are no data for Tajikistan. A single survey in Turkmenistan variously reported as providing data for 1990 or 1991 finds prevalence rates in men are lower than those elsewhere at 27%. This may reflect the fact that about one fifth of men use oral tobacco known as “nas”, a moist tobacco powder mixture, which is placed under the tongue. Officials in Turkmenistan report that when the use of “nas” is taken into account, the proportion of the adult population using tobacco products amounts to around 34%.126 In Kyrgyzstan the most recent data are limited to the capital Bishkek and suggest 60% of men and 12% of women smoke. Similar rates of male smoking are reported in Kazakhstan although there is uncertainty about the quality of data in women. In Uzbekistan, data suggest that 50% of men but very few women smoke although no recent data are available.

Data for the other countries are presented in Table 1-9-Table 1-14). Despite concerns about some of these data, a number of issues are apparent. First, men in the FSU have very high rates of smoking (between 45% and 60%). Such rates are amongst the highest reported anywhere in the world.118 Second, smoking prevalence in women is far lower than in men and lower than that seen in western Europe.118 Third, smoking amongst women in Estonia has historically been higher than among women elsewhere in the FSU. This finding is also reflected in a 1980 survey of school teachers in Estonia, where women of Russian nationality had significantly lower smoking rates (5.2%) than native Estonian women (13.3%). 127 These higher smoking rates are, however, now being matched by those in the other Baltic states and by women in urban areas elsewhere, for example in Ukraine. Fourth, in all countries where it was assessed, namely, Russia,128 Ukraine,129 Belarus130, Latvia and Lithuania,131 smoking among women in cities was higher than in those living in rural areas. This pattern was not seen in Estonia;131 in fact in the latest study there, the reverse was seen (Table 1-12). Trends over time can be best assessed in the Baltic republics, Russia, Ukraine and Belarus. A few tentative patterns can be ascertained (see Table 1-9-Table 1-14):

• Estonia - the prevalence increased until the mid 1990s in both sexes. Small decreases have since been seen,117,132 although the only formal analysis of trends over time indicates that, between 1994 and 1998, the downturn did not reach significance in

56 women and although significant in men, this was no longer so after adjusting for age, education and urbanisation.116

• Latvia – smoking increased in both sexes from 1994 to 1997 but has since appeared to stabilise.

• Lithuania – a steady increase among women from very low levels in the 1980s to reach 16% by 2000. Male data suggest an increase from the 1980s to early 1990s and smaller increases through the 1990s. National trends from 1994 to 1998 show the increase was significant in women but not in men once confounders were adjusted for116 and nor were trends in rural areas over a longer time period.133

• Belarus – stable in men, increasing gradually in women.

• Ukraine – there is some difficulty in assessing trends given the switch from urban to rural and then national samples. Nevertheless the data suggest a marked increase in women with particularly high rates now seen among those living in the capital, Kiev.

• Russia – a notable increase among men and women. The most accurate data are for Moscow and surveys over time indicate an increase in men from approximately 45% in the 1980s to 64% in 1996 and amongst women a tripling in prevalence from approximately 10% in the 1980s to 30% in 1996.

Further evidence of an increase among women comes from the pattern of age-specific rates in recent surveys which find fairly constant rates of smoking among men in successive generations but higher levels among younger women (Table 1-7-Table 1-14), suggesting that, for women, smoking is a relatively new phenomenon starting first amongst the youngest age groups.128,129,130,131 Increases in smoking prevalence are also suggested by the trends in youth smoking described below.

57 Table 1-7: Studies of smoking prevalence among adults in Armenia, Azerbaijan, Georgia and Moldova Study Geographic Year< NAge group Response Definition, Current smoking age, educational or notes area Rate (%) current rates (%) socioeconomic smoker Male Female differences Orlovski quoted in Armenia 1960 1685 n/a n/a n/a 77.9 6.9 n/a Zaridze WHO HFA database Armenia 1998 n/a 15+ n/a daily smoking 63.7 1.2 n/a WHO Global NCD Armenia 2000- 8149 15-54 na/ n/a 67.5 3.1 prevalence highest in Infobase 2001 middle-aged groups in both sexes

Orlovski quoted in Azerbaijan 1963 2079 n/a n/a n/a 68.6 4.8 n/a Zaridze TCCP database Azerbaijan 1999 n/a n/a n/a n/a 30.2 1.1 n/a Orlovski quoted in Georgia 1960 1238 n/a n/a n/a 78.5 7.5 n/a Zaridze Grim (1997) Georgia (Tbilisi) 1995 300 40-65 n/a n/a 57 13 n/a

WHO HFA database Georgia 1998 n/a adults n/a daily smoking 53.2 11.9 n/a WHO have no further details on the sample source (exact age (personal correspondence D R Prochorskas) uncertain) Grim (1999) quoted in Georgia (Tbilisi) 1999 321 40-69 85% at smoking now 60 15 n/a Sample only contained 92 men and 229 women TCCP & Global NCD (92 household because taken mainly during the day. Amongst Infobase men, level men and women combined, 33% smoked less 229 than 1 pack per day, 47% smoked 1-2 packs and wome 10% more than 2 packs per day. n)

Ioffina (additional data Moldova (1995) n/a n/a n/a n/a 46 7 n/a from Forey) TCCP database Moldova 1998 43.9 3 WHO HFA database Moldova 2000 40 2 WHO HFA database Moldova 2001 38.8 2 Notes

Sources used Forey B, Hamling J, Lee P, Wald N. International Smoking Statistics (second edition). Oxford: Oxford University Press, 2002. Grim CE, Grim CM, Kipshidze N, Kipshidze NN, Petersen J. DVD risk factors in Eastern Europe; a rapid survey of the capital of the Republic of Georgia [abstract]. Am J Hypertens

58 1997;10:211A Grim CE, Grim CM, Petersen JR, Li J, Tavill F, Kipshidze NN, Chawla PS. Prevalence of cardiovascular risk factors in the republic of Georgia. J Hum Hypertens,1999;13:243-247 Ioffina O, Kharchenko V, Akopian A, Rol' I znachenie tabakokureniia v zabolevaemosti I smertnosti ot boleznei sistemy krovoobrashcheniia v sovremennoi Rossii (Role of tobacco smoking on cardiovascular morbidity and mortality in today's Russia). Ter Arkh 1999; 71: 69-73. TCD database: The WHO Tobacco Control Database (http://cisid.who.dk/tobacco/). TCD source for Moldova, 1999: National Tobacco Agency TCCP database: Tobacco Control Country Profile database (http://www5.who.int/tobacco/page.cfm?sid=57. Source quoted for Azerbaijan and Moldova: household budget surveys. WHO Tobacco or Health: A global status report. http://www.cdc.gov/tobacco/who/whofirst.htm (last accesed 22/4/03) World Health Organisation Regional Office for Europe, Health for all database, updated January 2003 (http://hfadb.who.dk/hfa/) WHO Global NCD Infobase. Preliminary country reports available at: ftp://ftp.who.int/data/NMH/NCD-Surveillance/SuRF_1/ (last accessed 27 Oct 2003) Zaridze D. Dvoirin VV, Kobljakov VA, Pisklov VP. Smoking patterns in the USSR. In: Zaridze DG, Peto R (eds). Tobacco: A major international health hazard. IARC Scientific publications NO. 74. Lyon: IARC, 1986.

59 Table 1-8: Studies of smoking prevalence among adults in central Asia Study Geographic Year< N Age Response Definition, Current smoking rates (%) age, educational notes area group Rate (%) current or socioeconomic smoker Male Female differences

Ioffina (additional data Kazakhstan (1995) n/a 15+ n/a n/a 54 17 n/a from Forey) WHO HFA database Kazakhstan 1996 n/a 15+ n/a 60 7 n/a

WHO quoted in Forey Kyrgyzstan 1978 n/a 40-59 n/a n/a 52(46% among n/a n/a Kyrgyz and 54% among Russians)

WHO quoted in Forey Kyrgyzstan 1979 n/a 20-29 n/a daily smoking 72 (age 20-29) 15 (age 20- n/a (Bishkek) & 50-59 48 (age 50-59) 29)

Burkenkov quoted in Kyrgyzstan 1982 n/a 40-60 n/a daily smoking 48 n/a n/a Forey (Bishkek, referred to in paper as Frunze) WHO HFA database Kyrgyzstan 1997 2012 15-64 67% daily smoking 60 12 n/a WHO HFA database provides identical data for and personal (Polyclinics in 1997 and 1999. National experts suggest that correspondence Bishkek) the only survey undertaken was in 1997 & was Chinara Bekbasarova a casual sample from polyclinics in Bishkek (personal correspondence C Bekbasarova). WHO unable to provide further expla

60

Table cont'd WHO survey quoted in Turkmenistan (1991) 2676 n/a n/a n/a 27 0.5 n/a Studies in 1993 suggested that about 12% of Forey (Dashkhovuz) the population use "nas" (smokeless tobacco). Officials in Turkmenistan report that when the use of "nas" is taken into account, the proportion of the adult population using tobacco products amounts to around 34

Burkenkov quoted in Uzbekistan 1981 n/a 40-59 n/a daily smoking 56 n/a n/a Forey (Tashkent) Makhmudov quoted in Uzbekistan (1981) 1590 40-59 n/a n/a 51 n/a increased with Forey (Tashkent) declining age WHO quoted in Forey Uzbekistan 1989 n/a n/a n/a at least 1 40 in Uzbeks, 1 in Uzbeks, n/a cig/week 50 in other 3 in other population population groups groups

Piha a Uzbekistan 1991 n/a n/a n/a n/a 40 1 n/a WHO quoted in Forey Uzbekistan (1993) n/a n/a n/a n/a 49 9 increased with The largest percentage of smokers was found declining age in the age group 20-29 years (60% of men and 11% of women), while in the age group 30-39 years the corresponding figures were 53% and 15%. Increases in smoking in the youngest age groups during the last 20-30 ye

Ioffina (additional data Uzbekistan (1995) n/a n/a n/a n/a 47 2 n/a from Forey) (Tashkent)

61 Notes

Sources used Forey B, Hamling J, Lee P, Wald N. International Smoking Statistics (second edition). Oxford: Oxford University Press, 2002. Ioffina O, Kharchenko V, Akopian A, Rol' I znachenie tabakokureniia v zabolevaemosti I smertnosti ot boleznei sistemy krovoobrashcheniia v sovremennoi Rossii (Role of tobacco smoking on cardiovascular morbidity and mortality in today's Russia). Ter Arkh 1999; 71: 69-73. Piha T, Besselink E, Lopez AD. Tobacco or health. World Health Stat.Q. 1993;46:188-94. WHO Tobacco or Health: A global status report. http://www.cdc.gov/tobacco/who/whofirst.htm (last accesed 22/4/03) World Health Organisation Regional Office for Europe, Health for all database, updated January 2003 (http://hfadb.who.dk/hfa/)

62 Table 1-9: Studies of smoking prevalence among adults in Russia Study Geographic Year< n Age Response Definition, Current smoking age, educational or Desire to Urban rural Notes area group Rate (%) current rates (%) socioeconomic quit differences smoker differences Male Female Male Female

Oleynikov, Moscow 1975 2676 16+ 60 n/a 44.2 10.1 In both sexes prevalence n/a n/a n/a 1981 quoted (households) highest in those aged in Forey under 40 although prevalence in 16-20 years olds is low.

Oleynikov Moscow 1978 3983 40-59 66% n/a 48.2 n/a Smoking declined with n/a n/a n/a 1983 age and varied (additional significantly and inversely data from with level of education Forey) and employment

Chasova Moscow 1979 n/a 20-59 n/a n/a 54 13 In both sexes prevalence n/a n/a n/a data based on sample of highest in those aged those attending a clinic in under 40 Moscow. Shevtshuk Lipetsk 1981 1529 14-60 n/a smoking at 54 (66 in 6 (8 in Smoking less common age group given here is least 1 those those amongst university correct. Quoted as 12+ in cigarette per 18+) 16+) students than those in Forey which differs to day technical or vocational original paper training colleges Chasova Moscow 1983 n/a 20-59 n/a n/a 53 14 In both sexes prevalence data based on sample of highest in those aged those attending a clinic in under 40 Moscow. Centre for 20 cities and 1985,6 12,275 schoolchild n/a 45.1 Preventive six regions ren, Medicine, university reported in students & Tkachenko adults 20- 60

63 Table Cont'd Piha * Russia 1989 n/a n/a n/a n/a 44 10 n/a n/a n/a n/a

Dobson, Moscow 1984- 1175 35-64 70regular daily 40 13 n/a n/a n/a n/a Molaris (the Intervention 85 smokers, MONICA 1992- 1396 35-64 76occasional 42 14 n/a n/a n/a n/a study) 95 smokers (<1 per day) Moscow 1984- 1415 35-65 78 48 12 n/a n/a n/a n/a excluded Control 86 1992- 1084 35-65 66 47 14 n/a n/a n/a n/a 95 Novosibirsk 1985- 1267 35-65 73 54 3 n/a n/a n/a n/a intervention 86 1994- 1273 35-65 73 58 8 n/a n/a n/a n/a 95 Novosibirsk 1985- 1178 35-65 71 61 4 n/a n/a n/a n/a control 86 1995 1175 35-65 70 60 6 n/a n/a Chasova Moscow 1986 n/a 20-59 n/a n/a 49 9 In both sexes prevalence n/a n/a n/a data based on sample of highest in those aged those attending a clinic in under 40 Moscow. Elgarov Nalchik,southe (1991) 2562 20-54 70 daily 55 n/a Smoking more common n/a n/a n/a rn Russia smoking in manual than non- manual workers WHO HFA Russia 1991 n/a 15+ n/a daily 47 12 n/a n/a n/a n/a db smoking

64 Table Cont'd Laatikainen Pitkaranta, 1992 1000 25-64 83.5 regular daily 65 10$ Smoking more common 76% men, n/a n/a $use of blood cotinine Republic of smokers, in lower educational 71% women levels showed women Karelia, (ural occasional groups in men but wished to under-reported smoking. If area of North- smokers (<1 differences not quit but 62% high cotinine levels used, West Russia) per day) significant.Prevalence & 56% smoking prevalence would excluded increased with decreasing respectively rise to 21% in women and age in both sexes. had never 68% in men (suggests attempted to cultural unacceptability of do so female smoking)

Trubacheva Small towns in (1993) 10836 20-59 n/a daily 69 6 n/a n/a n/a n/a nb this is probably 1992 (additional western smoking sample in Siberia quoted by data from Siberia Hurt with almost identical Forey) results

Bilitchenko Moscow 1993 n/a 25-65 n/a n/a 46 12 n/a n/a n/a n/a Pakriev Udmurtia 1995 855 18-65 n/a n/a 68.8 2.3 No significant educational 60% of ever n/a n/a majority smoked cheap low (largely rural, differences. Smoking smokers quality cigarettes without low-income significantly higher had filters. Smoking associated area) amongst workers than attempted to with alcohol dependence employed and amongst quit, but only (75% of those with alcohol ethnic Russian men than 5% dependence were tobacco Urdmurts abstinent for users) and suicide attempts > 1 year. in men. Mean age of onset 19.7 years. Among men 52.7% were tobacco depe

RLMS round Russia 1995 4000 14+ 80% hhs, n/a 59 9 n/a n/a n/a n/a 6 quoted in househo 97% adults Forey lds

65 Table Cont'd Levshin Moscow 1995-7 3000 14-19 & n/a n/a 44, 59 13, 18 In men smoking higher in n/a n/a Rates among doctors were (additional 25-69 less well educated as high and in women data from higher than among the Forey) general population.

McKee Russia 1996 1587 18+ 66 all current 63 14 No clear educational n/a no significantl smokers differences. Smoking differenc y higher in Moscow 1996n/a by 18+ n/a by area all current 64 30 significantly more e urban area smokers common in deprived areas groups. In men more Rural 1996 18+ n/a by area all current 62 9 common in middle aged smokers men and in women in youngest compared with older age groups

Shalnova 1998 63 10 quoted in TCD & WHO HFA db

Moscow Moscow 2000- 1689 25-64 n/a daily 57 21.4 in women highest in n/a n/a n/a behaviour 2001 smoking youngest age groups risk factor survey quoted in WHO Global NCD Infobase

Sources used Bilitchenko TN, Kharlova NV. Smoking prevention politics in Russia [Abstract]. Conference on Global lung health and the 1996 annual meeting of the International Union Against Tuberculosis and Lung Disease. Tuber Lung Dis 1996;77(Suppl 2):42. Chasova LV, Prokhorova IN, Prokhorov AV, Ivanov VM et al. Smoking and some measures of its control Ter Arkh 1989;1:40-3. Dobson AJ, Evans A, Ferrario M, Kuulasmaa KA, Moltchanov, VA et al . Changes in estimated coronary risk in the 1980s: data from 38 populations in the WHO MONICA Project. World Health Organization. Monitoring trends and determinants in cardiovascular diseases. Annals of Medicine 1998;30:199-205. Elgarov AA, Zhasminova VG, Konstantinov VV, Sokolva MA, Khukhokova MM. Chastota ishemicheskoi bolezni serdtsa I faktorov ee riska u muzhchin 20-54 let-rabotnikof fizicheskogo i umstvennogo truda (The incidence of coronary heart diesease and its risk factors in men aged 20-54 years, blue and white collar workers.) 66 Ter Ark 1992;64:100-3. Forey B, Hamling J, Lee P, Wald N. International Smoking Statistics (second edition). Oxford: Oxford University Press, 2002. Laatikainen T, Vartiainen E, Puska P. Comparing smoking and smoking cessation process in the Republic of Karelia, Russia and North Karelia, Finland. Journal of Epidemiology & Community Health 1999;53:528-34. Levshin V, Drojachih V, Fedichkina T, Slepchenko N. Epidemioloy of smoking and some determinants of smoking behaviour. In: Lu R, Mackay, J, Niu S and Peto R (eds) Tobacco: the growing epidemic. Proceedings of the Tenth World Conference on Tobacco or Health, 24-28 August 1997, Beijing,China. London: Sringer-Verlag, 2000:52-4. McKee M, Bobak M, Rose R, Shkolnikov V, Chenet L, Leon et al . Patterns of smoking in Russia. Tobacco Control 1998;7:22-6. Molarius A, Parsons RW, Dobson AJ, Evans A, Fortmann SP, Jamrozik K et al . Trends in cigarette smoking in 36 populations from the early 1980s to the mid-1990s: findings from the WHO MONICA Project. Am.J.Public Health 2001; 91:206-12. Oleynikov SP, Chasova LU, Glazunov . Smoking and some socio-demographic characteristics: data from research in Moscow and Kaunas. Ter Arch 1983;1 :57-61 Pakreiv S, Shlik J. Patterns of tobacco use in rural Udmurtia. Tobacco Control 2001 10: 85-86 Pakreiv S, Shlik J. Patterns of tobacco use in rural Udmurtia, 2001. unpublished report. Piha T, Besselink E, Lopez AD. Tobacco or health. World Health Stat.Q. 1993;46:188-94. Shevtshuk AG, Tarasova RN. Ob organizatsii sanitarno-prosvetitelnoy raboti v borbe c kureniem. [Organisation of an anti-smoking campaign). Zdravookhranenie Rossiiskoi Federatsii 1983;5:27-8. TCD database: The WHO Tobacco Control Database (http://cisid.who.dk/tobacco/). Tkachenko GB, Prokhorov A. Tobacco and Healthin the Russian Federation. Copenhagen: WHO Regional Office for Europe. (EUR/ICP/LVNG 030101), 1997. Trubacheva IA, Levitsky EF, Perminova OA. Epidemiology ofrisk factors for coronary athersclerosis in populations of small towns of western Siberia [Abstract] Atherosclerosis 1994; 109:20-1. World Health Organisation Regional Office for Europe, Health for all database, updated January 2003 (http://hfadb.who.dk/hfa/)

Original sources quoted but not seen Shalnova, S.A., et al. [Prevalence of smoking in Russia. Results of a survey of a nationally representative population sample.] Profilaktika zabolevanij i ukreplenie zdorov’ja, 3 (1998)

Other sources not used Hurt, R. Smoking in Russia: what to Stalin and western tobacco companies have in common? Mayo Clin Proc 70, 1007-1011. 1995. This quoted rates as: 53% in 1985 increasing to 67% in 1992 in men. In industrial areas in 1986: 78.6% in men, 13.7% in women, in Moscow 49% in men and 9.2% in women. In siberia in 1992: 69.4% in men and 7.9% in women The data Hurt quotes come from :Prokharov AV, Smoking in Russia. WHO Regional Office for Euope, Copenhagen 1993 and Tchachenko GB, Riazantsev VK. Description of the situation in the Russian Federation. WHO Regional Office for Europe, Copenhagen 1993.

67 Table 1-10: Studies of smoking prevalence among adults in Ukraine Study Geographic Year< N Age Response Definition current Current smoking Age, educational or socioeconomic Urban rural Notes area group Rate (%) smoker rates (%) differences differences

Male Female Male Female ADIC 1 ?Kiev (not 1977-8 n/a 20-59 n/a n/a 51.1 5.2 Smoking most common in younger age n/a all thought to be (Institute of stated) groups although differences greater in urban samples Cardiology) 1982-3 n/a 20-59 n/a n/a 50.6 5 women.(significance not assessed) n/a although only 1987-8 n/a 20-59 n/a n/a 49.9 8.3 Higher rates of smoking amongst women n/a one clearly with higher education compared with stated as such Kiev 1994-5* n/a 20-59 n/a n/a 48.5 20.5 n/a those with primary education. Amongst 1998-9 n/a 20-59 n/a n/a 53.3 n/a n/a men, most common in those with sec Gorbas Ukraine - (1981) 2191 20-59 n/a n/a 54.1 5.3 n/a n/a n/a Rates in women (1994) urban increased population significantly over time (p<0.001)

(1993) 1901 20-59 n/a n/a 48.9 11.2 n/a n/a n/a

Shuteeva Kiev 1990 1500 20-69 n/a >1 cig/day for > 1 51.1 n/a Smoking highest in youngest age groups n/a n/a year + those who had smoked this amount but stopped < 1 year before study

Gorbas rural n/a (pre 3271 n/a n/a n/a 60.7 0.5-0.6 n/a Yes: more Yes: 10 population 1996) common in times rural more dwellers common in urban areas Ioffina n/a (1995) n/a n/a n/a n/a 60 12 n/a n/a n/a

68 Table cont'd Institute of Not stated - 1999 n/a 20-59 n/a n/a 51.1 19.4 n/a n/a n/a Sample data not Cardiology, thought to be provided quoted in urban although thought TCD to be urban sample (personal correspondence K Krasovsky)

Gilmore Ukraine 2000 1590 18+ 72 all current smokers 57 10 Significantly higher smoking rates No Yes: amongst deprived and unemployed of smoking both genders. No significant difference by rates over income or education 7 times higher in the most urban areas

ADIC Ukraine 2000 1797 15+ n/a Daily smoking 58 14 Prevalence was greater in university than 2(State primary educated men but the opposite Statistical pattern was seen in women although the Committee) proportion of quitters did increase with ¬ education. Smoking in women highest in yougest and in men in middle age groups.

69 Table cont'd ADIC 2 Ukraine 2001 2721 14+ n/a Daily smoking 56.9 14.2 In men prevalence was highest in those (Ukrainian aged 25-39, no clear pattern was seen Institute of with education or material status although Social prevalence was highest in those with Research lowest status (significance not assessed). data) In women smoking was highest in those aged 18-24, decli

Sources used ADIC 1: http://adic-co.info.kiev.ua/toborhealth/prevalence.htm (last accessed 4/9/01) ADIC 2: Alcohol and Drug Information Center (ADIC- Ukraine). Economics of tobacco control in Ukraine from the public health perspective. Kiev: ADIC, 2002. http://www.adic.org.ua/adic/reports/econ/ch-3/3-3.htm (last accessed 17/04/03). Forey B, Hamling J, Lee P, Wald N. International Smoking Statistics (second edition). Oxford: Oxford University Press, 2002. Gilmore A, McKee M, Telishevska M, Rose R. Smoking in Ukraine: epidemiology and determinants. Preventive Medicine 2001;33:453-61. Gorbas I, Smirnova I, vasha E, Davidenko N, Koblyanskaya AV. Dynamics of the epidemiological situation coronary heart disease and risk factors within the urban population of Ukraine.[Abstract] Eur Heart J 1996; 15(suppl): 373. Gorbas, I. Cigarette smoking prevalence in Ukraine rural population in connection with cardiovascular diseases generation. Paper presented at the Smoke Free Europe Conference. 1996 Ioffina O, Kharchenko V, Akopian A, Rol' I znachenie tabakokureniia v zabolevaemosti I smertnosti ot boleznei sistemy krovoobrashcheniia v sovremennoi Rossii . (Role of tobacco smoking on cardiovascular morbidity and mortality in today's Russia). Ter Arkh 1999; 71: 69-73 Shuteeva L. Epidemiological aspects of tobacco smoking in connection with its clnical course among men aged 20-69 years living in Kiev [Russian]. Ter Arkhiv 1990; 62(12):108-10. TCD database: The WHO Tobacco Control Database (http://cisid.who.dk/tobacco/).

Other sources not used in table Kvasha E, Smirnova I, Corba I, Daidenko N. Smoking among the Ukraine female population. In: Tobacco: the growing epidemic. Abstract book, 10th World Conference on Tobacco or Health. Beijing, China. 24-28 August 1997. 142 (PO46). - Female smoking rates were assessed in sample of 2259 women across a 5-year period. Survey years and age group are not stated (but estimated by Forey to be circa 1990 and 1995). The prevalence in women increases over the 5-year period from 5% to 21%. In addition it is stated that 99% of smoking women are urban citizens. Original sources quoted but not seen Institute of Cardiology, quoted in TCD

70 Table 1-11: Studies of smoking prevalence amongst adults in Belarus Study Geographic Year< N Age Response Definition, current Current smoking rates (%)age, educational or Urban rural area group Rate (%) smoker socioeconomic differences differences Male Female Male Female Burkenkov quoted in Forey Belarus 1981 n/a 40-59 n/a daily smoking 52.0 n/a n/a (Minsk) WHO HFA db Belarus 1995 n/a 15+ n/a daily smoking 54.8 3.6 WHO HFA db Belarus 1996 n/a 15+ n/a daily smoking 54.7 4.6 WHO HFA db Belarus 1997 n/a 15+ n/a daily smoking 53.4 4.8 WHO HFA db Belarus 1998 n/a 15+ n/a daily smoking 54.9 4.6 WHO HFA db Belarus 1999 n/a 15+ n/a daily smoking 53.7 4.8 WHO HFA db Belarus 2000 n/a 15+ n/a daily smoking 54.0 6.7 Gilmore Belarus 2000 1087 18+ 53.4 daily smoking 53.4 9.4 Very significant increase in smoking No Significant in younger compared with older urban bias in women. Men with lower education women and material disadvantage more than those who are better off.

WHO HFA db Belarus 2001 n/a 15+ n/a daily smoking 53.3 6.3

Sources used Forey B, Hamling J, Lee P, Wald N. International Smoking Statistics (second edition). Oxford: Oxford University Press, 2002. Gilmore A, McKee M, Rose R. Smoking in Belarus: evidence from a household survey. European Journal of Epidemiology 2001;17:245-53 World Health Organisation Regional Office for Europe, Health for all database, updated January 2003 (http://hfadb.who.dk/hfa/)

71 Table 1-12: Studies of smoking prevalence among adults in Estonia Study Geographic Year N Age Response Definition, Current smoking Age, educational or Urban rural differences Notes area group Rate (%) current rates (%) socioeconomic differences smoker Male Female Male Female Volozh Tallinn1981-2 217730-59 70.2 regular daily n/a n/a n/a n/a n/a 1984-8 1893 71.8smoking (>1 n/a n/a n/a n/a n/a 1992-4 949 50.6per day) n/a n/a n/a n/a n/a

Raudsepp Estonia 1980 8347 teachers 82 anyone 40.3 11.1 Among women, smoking more n/a n/a data on knowledge of (school aged 29- considering common in youngest age groups health impact of teachers only) 60 themselves and among Estonian than Russian smoking given to be a women smoker

Pudule Estonia 1997 2019 20-64 67.3 Daily 53.9 24.1 Amongst men, smoking more not seen not seen smoking common in those aged 35 - 49 and of low income (significant) & education (non-significant). Smoking in women far less common in oldest than two younger age groups although differences not significant.

72 Table cont'd Kasmel et al Estonia1990 1085 18-64 72 Daily 46 (48) 15 (17) Based on analysis of 1994-1998 Puska paper: (additional 1992 941 16-64 63(regular) 49 (52) 20 (22) surveys: Smoking significantly less Downward trend seen data from 1994 1243 16-64 83smoking 52 (55) 23 (25) common in the oldest compared not seen not seen but did not reach Prattala R et 1996 1507 16-64 75 47 (50) 22 (23) with youngest age group and in most significance in women. al, Puska et compared with least educated. In In men was significant 1998 1362 16-64 68 42 (44) 20 (21) al and 2002 survey significance not but not after adjusting 2000 1376 16-64 68 44 (47) 20 (21) personal assessed but smoking rates in men for age, education and 2002 1338 16-64 67 45 (46) 18 (19) highest in rural areas and corresponden highest in middle age. In urbanisation. 2003 falls with urbanicity to ce) report states that daily lowest level in Tallinn smoking decreasing (significance not assessed) since 1996.

* Heavy smokers defined here as smoking >20 cigarettes per day

Sources used: Kasmel A, Lipand A, Markina A. Health behaviour among Estonian adult population, Spring 2002. Estonian Centre for Health Education and Promotion, Tallinn, 2003. Pudule I, Grinberga D, Kadziauskiene K, Abaravicius A, Vaask S, Robertson A et al . Patterns of smoking in the Baltic Republics. Journal of Epidemiology & Community Health 1999;53:277-82. Puska P, Helasoja V, Prattala R, Kasmel A, Klumbiene J. Health behavour in Estonia, Finland and Lithuania 1994-1999. European Journal of Public Health 2003; 13: 11-17. Prattala R, Helasoja V, the Finbalt Group. Finbalt Health Monitor: feasibility of a collaborative system for monitoring health behaviour in Finland and the Baltic countries . Helsinki: National Public Health Institute 1999. Raudsepp J,.Rahu M. Smoking among schoolteachers in Estonia 1980. Scand.J.Soc.Med. 1984; 12:49-53. Volosh O, Deev A, Solodkaya E, Abina J, Kalyuste T, Kaup R et al. Assessment of the general health profile trends in the male population of Tallinn, Estonia. Public Health 1998; 112: 303-308.

73 Table 1-13: Studies of smoking prevalence among adults in Latvia Study Geographic Year N Age Response Definition, Current smoking rates Age, educational or socioeconomic differences Urban rural differences area group rate current Male(%) Female Male Female smoker WHO HFA db Latvia 1994 n/a 15+ n/a Daily smoking 49.4 11.1 n/a n/a n/a

Pudule 1999 Latvia 1997 2331 19-64 77.7 Daily smoking 56 11 Amongst men, smoking more common in those not seen Significantly less (Baltic Nutrition aged 35 - 49 and of low income (significant) & common in rural survey)* education (non-significant). Smoking in women far areas less common in oldest than two younger age groups although differences not significant.

Pudule 1999 Latvia 1998 2318 15-64 77.3 Daily smoking 51.3 19.2 (Finbalt data)**

Pudule 2001 Latvia 2000 2400 15-64 80 Daily smoking 51.3 18.2 (Finbalt data) Pudule 2003 Latvia 2002 2029 15-64 67.6 Daily smoking 51.1 19.2 In both genders, smoking varies inversely with rates similar rates similar (Finbalt data) education and social class (lowest rates in those with (significance (significance not higher education and social status). In women not tested) tested) smoking lowest in oldest age groups, then increases with declining age although little difference between t

*Data in Nutrition survey and Finbalt Surveys were collected in different ways and the author suspects this may account for the different prevalence rates in women. The former used interviewer administered questionnaires and the latter self completed postal questionnaires. Pudule suggests that differences in prevalence in women may be accounted for in part by women not wishing to disclose their smoking behaviour, still relatively taboo, to interviewers (personal correspondece I Pudule May 2003). Thus female smoking rate in 1997 likely to be somewhere between 11 and 19%. ** Data from this survey is also provided in the WHO HFA database and the TCCP database. However the author has since re-analysed the data treating mssing values differently and suggests the data presented here should be used. They have the advantage of being compatable with the 2000 and 2002 data. Sources used: Pudule I, Grinberga D, Kadziauskiene K, Abaravicius A, Vaask S, Robertson A et al . Patterns of smoking in the Baltic Republics. Journal of Epidemiology & Community Health 1999;53:277-82.

74 Pudule I, Grīnberga D, Rituma A, Villeruša A, Zīle S, Prättälä R, Helasoja V, Puska P “Latvijas iedzīvotāju veselību ietekmējošo paradumu pētījums, 1998.”, (Health Behaviour among Latvian Adult Population, 1998), Helsinki: National Public Health Institute, 2000. With additional data on sample size and response rates from Prattala R, Helasoja V, the Finbalt Group. Finbalt Health Monitor: feasibility of a collaborative system for monitoring health behaviour in Finland and the Baltic countries. Helsinki: National Public Health Institute 1999. Pudule I, Grīnberga D, Rituma A, Villeruša A, Zīle S, Prättälä R, Helasoja V, Puska P “Latvijas iedzīvotāju veselību ietekmējošo paradumu pētījums, 2000.”, (Health Behaviour among Latvian Adult Population, 2000), Helsinki: National Public Health Institute, 2001 Pudule I, Grīnberga D, Villeruša A, Dzerve V, Zīle S, Helasoja V, Vähäsarja K, Prättälä R. “Latvijas iedzīvotāju veselību ietekmējošo paradumu pētījums, 2002.”, (Health Behaviour among Latvian Adult Population, 2002), Helsinki: National Public Health Institute, 2003 Pudule I (personal correspondence), May 2003

75

Table 1-14: Studies of smoking prevalence among adults in Lithuania Study Geographic Year n Age Response Definition, Current smoking Age, educational or socioeconomic Urban rural differences Notes area group Rate (%) current rates (%) differences smoker Male Female Male Female Oleynikov 1983 Kaunas 1978 5482 40-59 69.2 n/a 43.1 n/a Smoking declined with age varied significantly and inversely with level of education and employment

Molarius, the Kaunas 1983- 1463 35-64 69 Daily 38 4 n/a n/a n/a MONICA study 85 smoking

Piha (WHO) ** Lithuania 1992 n/a n/a n/a n/a 52 10 n/a n/a n/a

Molarius, the Kaunas 1992- 1239 35-64 76 Daily 35 4 n/a n/a n/a MONICA study 93 smoking

Puska et al* Lithuania 1994 1864 20-64 62 Daily 44 6significantly higher in youngest age not lower in rural Trends over time smoking groups especially in women,in those significant areas but assessed - with least education in men but those only significant increases with medium education in women. borderline seen in men and significance women but after adjusting for age, education and urbanisation remained significant only in women (p<0.001)

1996 2021 20-64 67 Daily 48 10 smoking 1998 1874 20-64 62 Daily 51 13 smoking Pudule * Lithuania 1997 2223 20-64 74.1 Daily 53.2 7.6 Amongst men, smoking decreases not seen Significantly smoking witn increasing income (significant) & less common education (non-significant). Smoking in in rural areas women far less common in oldest than two younger age groups (significant)

76

Table cont'd TCCP db (from Lithuania 1997 n/a 18+ n/a Regular 41.4 8.6 n/a n/a n/a National Centre smoking for Health Promotion and Education)

Plieskiene Lithuania (rural 1987, 5140 25-64 n/a n/a 46.5 in 3.7% in smoking more common in younger n/a n/a smoking also areas) 1993, across 1999, 1987, age groups in both sexes, inverse associated with 1999 the 3 said not 12.6% in relationship with education only in alcohol consumption surveys to have 1999 men. especially in women change d across WHO HFA Lithuania 2000 n/a 20-64 n/a Daily 51.5 15.8 n/a n/a n/a data smoking *Pudule study (based on the Baltic Nutrition Study ) and Puska study (FINBALT surveys) collected data in different ways. The former used interviewer administered questionnaires and the latter self completed postal questionnaires. Pudule suggests that differences in prevalence in women may be accounted for by women not wishing to disclose their smoking behaviour, still relatively taboo, to interviewers (personal correspondece I Pudule May 2003). WHO HFA database gives identical or almost identical data fo 1994, 1996 and 1998 Sources used Piha T, Besselink E, Lopez AD. Tobacco or health. World Health Stat.Q. 1993;46:188-94. Plieskiene A, Milasauskiene z, Klumbiene J. Relationship between smoking habits and other riskfactors in Lithuanian rural population. Submitted to 3rd European Conference on Tobacco or Health, Warsaw 2002. Pudule I, Grinberga D, Kadziauskiene K, Abaravicius A, Vaask S, Robertson A et al . Patterns of smoking in the Baltic Republics. Journal of Epidemiology & Community Health 1999;53:277-82. Puska P, Helasoja V, Prattala R, Kasmel A, Klumbiene J. Health behavour in Estonia, Finland and Lithuania 1994-1999. European Journal of Publif Health 2003; 13: 11-17. Additional data (on sample size and response rates from Prattala R, Helasoja V, the FInbalt Group. FInbalt Health Monitor: feasibility of a collaborative system for monitoring health behaviour in Finland and the Baltic countries. Helsinki: National PUblic Health Institute 1999. Molarius A, Parsons RW, Dobson AJ, Evans A, Fortmann SP, Jamrozik K et al . Trends in cigarette smoking in 36 populations from the early 1980s to the mid-1990s: findings from the WHO MONICA Project. Am.J.Public Health 2001; 91:206-12. Oleynikov SP, Chasova LU, Glazunov . Smoking and some socio-demographic characteristics: data from research in Moscow and Kaunas. Ter Arch 1983;1 :57-61 TCCP database (http://www5.who.int/tobacco/page.cfm?sid=57) WHO HFA database (http://hfadb.who.dk/hfa/)

NOTE FOR ALL TABLES ABOVE: Green text used for smoking prevalence rates when the sample is thought to be urban but is not quoted as such (samples quoted as urban are left in normal text)

77 3.4 Smoking prevalence - youth

Three major international youth smoking surveys are conducted regularly and include the three Baltic republics, Russia and Ukraine. These are the Health Behaviour in School- Aged Children (HBSC),134 the European School Survey Project on Alcohol and Other Drugs (ESPAD)135 and the Global Youth Tobacco Survey (GYTS)136 (Table 1-15).

These surveys all collect data on experimental (e.g. ever smoked or cigarette use during last 30 days) and regular smoking (weekly or daily smoking and lifetime use over a certain amount). However, the exact questions, systems of data collection and age groups surveyed differ, making direct comparisons and analyses over time difficult. Nevertheless a number of issues are apparent from the FSU results, presented in Table 1-16-Table 1-18, accompanied by selected international comparisons.n

In the FSU, boys experiment more than girls and are more likely to become regular smokers. Outside the FSU and CEE (for which data is not shown), although boys often experiment more than girls, girls are at least equally as likely to become regular smokers and thus regular smoking tends to be more common in girls, or the genders have similar rates(Table 1-16,Table 1-17). The FSU (along with CEE) is therefore notable in terms of its wide gender differences and male predominance.

These findings impact on international comparisons – boys in the FSU have similar or higher smoking rates than their peers elsewhere, a pattern also observed in a 1990 study comparing the US (Minnesota) with Moscow.137 By contrast, rates among girls in the FSU have tended to be lower than their international counterparts134,137 although they are now becoming more similar.138

The HBSC survey suggests that, in younger age groups, experimental smoking is more common in the Baltic republics than in Russia (Table 1-16). However, in older age groups and established smokers these between-country differences reduce and by the age of 15, girls in Russia have higher regular smoking rates than those in the Baltic republics (Table 1-16). These findings are supported by the GYTS data, which show that ever smoking is more common outside Russia while current smoking is similar in the four countries surveyed (Table 1-18).

n Greenland because it has very high rates, Greece where rates tend to be low and the US or UK which have more average rates.

78 Perhaps for this reason (although an alternative explanation is the urban nature of the sample in Russia compared with national samples elsewhere) international comparisons show that by the age of 15 Russian youth have high smoking rates – the 3rd highest out of almost 30 European countries surveyed (Table 1-17). Latvia, Lithuania and Ukraine tend to have rates above or around the average while rates in Estonia are lower. Wider international comparison beyond Europe is possible via the GYTS data, which suggests that rates of ever and current smoking among young people in the FSU are considerably higher than those in a diverse group of mainly low and middle income countries.139

Over the 4 year period between HBSC surveys, rates of experimental and regular smoking have increased in both genders in almost all age groups, although not all changes were significant. In Lithuania, Latvia and Russia weekly smoking among 15 year olds boys showed a significant increase that was most marked in Lithuania. In girls the increase was significant in all four FSU countries but most notable in Lithuania and Russia. 138 Similarly, in the ESPAD surveys, the most notable increases were seen in Lithuania where daily cigarette use increased from 13% to 17% in boys and doubled in girls from 3% to 6%. Cigarette use in the last 30 days also showed marked increases, again almost doubling in girls from 18% to 30% and in boys increasing from 34% to 49% (Table 1-17).

Other evidence is consistent with the view that youth smoking rates in the region have been increasing. Repeat surveys in Moscow conducted by the State Centre for Preventive Medicine found that prevalence had increased between 1992 and 1993 from 22.9% to 27.2% among 11-14 year-old boys, from 40.5% to 48.6% among 13-16 year old boys, and from 11.5% to 14.9% and 24.5% to 37.7% among girls in the same age groups.73,140 The definitions of smoking used are not given but comparisons with other data suggest these must be rates of experimental rather than regular smoking. Published results from repeat surveys among school children in Moscow show marked increases in smoking in both genders between 1986 and 1995141,142 and the ESPAD data for Moscow suggest further increases since 1995 (Table 1-19).

The GYTS also provides information on exposure to tobacco promotions. In Russia, 17% of children had been offered free cigarettes, with lower rates in Ukraine, Latvia and Lithuania. Other than in Lithuania, pro-smoking messages were seen more frequently

79 than anti-smoking messages. To compound this problem, the proportion receiving health education in schools was low, particularly in Russia (Table 1-18).

80 Table 1-15 Details of youth smoking surveys conducted in the FSU HBSC EPSAD GYTS Countries Estonia Estonia Latvia Latvia Latvia Lithuania Lithuania Lithuania Russia Russia Russia Ukraine Ukraine Organisation WHO Swedish Council for CDC & WHO Information on Alcohol and other Drugs, Pompidou Group – Council of Europe Age group 11, 13 and 15 year 15-16 year olds 13 – 15 year olds olds (students that will become 16 during the year of data collection) Years of data 1993/4 1995 Various: 1999 to 2002 collection 1997/8 1999 4 – yearly thereafter Sample size 4500 (1500 from Approx 2000 to 5000 n/a each age group) for FSU countries included Sampling Cluster sampling of Cluster sampling of Multistage cluster methodology school classes school classes with sampling of school some variation in classes with selection probability of sampling proportional to enrolment size No. of countries 25 in 1993/4 26 in 1995 Approximately 140 included 28 in 1997/8 30 in 1999 countries Region covered Europe Europe Global

81 Table 1-16 HBSC data on experimental and regular youth smoking in 1993/4 (shown in brackets) and 1997/8 by sex, country and age group Ever smoked a cigarette (%) Weekly smokers (%) Daily smokers (%) 11-yr olds 13- yr olds 15-yr olds 11-yr olds 13- yr olds 15-yr olds 11-yr olds 13- yr olds 15-yr olds Male Female Male Female Male Female Male Female Male Female Male Female male female male female male female Estonia 47 (41) 12 (8) 68 (63) 36 (29) 81 (81) 56 (46) 1 (1) 1 (0.2) 7 (6) 2 (1) 24 (22) 12 (6) 0 1 4 1 17 8 Latvia 41 (48) 15 (13) 74 (68) 52 (32) 87 (81) 68 (55) 4 (1) 0.3 (1) 15 (13) 7 (3) 37 (33) 19 (14) 2 0.1 8 3 27 12 Lithuania 46 (31) 19 (6) 70 (58) 41 (21) 83 (77) 55 (41) 2 (1) 0.3 (0.1) 10 (3) 2 (0.2) 24 (15) 10 (4) 1 0.3 6 1 15 6 Russia+ 24 (29) 9 (9) 52 (51) 42 (29) 69 (60) 56 (44) 2 (3) 0.5 (0.3) 13 (13) 7 (5) 24 (19) 22 (10) 0.4 0 7 3 20 14 Greenland* 38 (31) 40 (40) 66 (59) 80 (74) 82 (76) 90 (84) 3 (4) 6 (6) 29 (17) 41(22) 52 (49) 63 (46) 2 2 19 29 45 56 Greece** 5 3 23 20 48 45 1 1 5 5 18 19 0.1 0.5 3 2 13 14 USA 22 16 42 42 61 60 3 2 8 9 20 21 2 1 5 3 13 12 *Greenland selected as it has the highest prevalence in 5 of the 6 categories, **Greece selected to illustrate the lower prevalence range (in bottom 5 for 5 of the 6 categories), + The Russian sample is drawn just from St Petersburg and district. Source: WHO, HBSC Surveys. 1993/4 data: http://www.hbsc.org/downloads/HealthofYouth1.pdf (last accessed 22/4/03), 1997/8 data: http://www.hbsc.org/downloads/Int_Report_00.pdf (last accessed 22/4/03)

Table 1-17 ESPAD data on youth smoking in 15-16 year olds (those born in 1983) in 1999

Cigarette use during last 30 days Lifetime use of > 40 cigarettes Daily smoking at age 13 or younger Male FemaleBoth Rank Male FemaleBoth Rank Male Female Both Rank of 28 of 28 of 29 1999 1995 1999 1995 1999 1999 1999 1995 1999 1995 1999 1999 1999 1995 1999 1995 1999 1999 Estonia41372422322238361817271612154 42721 Latvia 48 34 40 9 38 23 20 11 13 6 9 20 Lithuania4934301840104629231235717136 31111 Russia *484245346384231815163 Ukraine50512928401239411818291315145 41916 Greenland62716714155501920146 Greece343635192827271753 329 UK 31 32 47 40 34 20 24 25 28 30 26 20 16 15 24 22 20 1 *Moscow only. Source: Hibell, B, Andersson B, Ahlstrom S, Balakireva O, Bjarnasson T, Kokkevi A, Morgan M. The 1999 ESPAD Report. Alcohol and other drug use among students in 30 European countries. Stockholm: Modin Tryck AB.

82 Table 1-18 GYTS data on youth smoking patterns Year Grade/YearEver smoking Current smoking Saw print Saw Offered Have Saw anti media ads billboard free object with smoking adds cigarettes brand logo media messages Male Female Both Male Female Both in last 30 days Latvia 2002 grade 7-9 86.6 72.2 79.9 38.2 29.7 34.3 86.8 82.1 5.3 33.3 68.0 Lithuania 2001 grade 7-9 83.6 68.0 75.8 40.3 31.3 35.7 66.5 72.5 4.2 25.7 82.6 Russia (Moscow) 1999 grade 7-10 71.1 61.7 66.7 40.6 29.8 35.3 76.5 94.3 17 22.1 74.5 Ukraine (Kiev) 1999 grade 8-10 84 69.1 77.3 46.8 33.8 41.1 87.4 n/a 8.1 26.1 78.7 Source: WHO GYTS Fact Sheets http://www.cdc.gov/tobacco/global/gyts/GYTS_factsheets.htm (last accessed 16/5/03)

Table 1-19 Trends in youth smoking in Moscow Weekly smoking Daily smoking 13- year olds 15- year olds 13- year olds 15- year olds Boys Girls Boys Girls Boys Girls Boys Girls 1986 7 3 23 8 3 0 17 5 1995 18 1 38 21 1999 18 15 Sources: 1986 data - Prokhorov AV,.Alexandrov AA. Tobacco smoking in Moscow school students. Br.J.Addict. 1992;87:1469-76 with further details taken from Forey, 2002. 1995 data - Alexandrov AA, Alexandrova V, Ivanova EI. Prevalence of smoking among Moscow school children and approaches to prevention. In: Lu R, Mackay J, Niu S and Peto R (Eds). Tobacco: the growing epidemic. Procedings of the Tenth World Conference on Tobacco or Health, 24-28 August 1997, Beijing, China. London: Springer-Verlag, 266-8. Additional data from Forey, 2002. 1999 data - ESPAD survey (reports daily smoking at age 15-16) NB ESPAD data may not be entirely comparable with data in other 15 year olds as survey was conducted in 15 -16 year olds

83 3.5 Smoking amongst doctors

In most populations smoking amongst the medical profession is lower than amongst the general population. This pattern is not, however, generally seen in the FSU where high rates of smoking amongst clinicians are commonplace. A survey in Moscow performed in the mid 1990s found that the prevalence of smoking amongst male doctors was the same as that in the general population and the prevalence among female doctors even higher.143 A survey in the late 1990s found very high rates of smoking - 56% of male and 35% of female medical students smoked as did 50% of male and 27% of female student nurses. 67% wanted to quit and most had already tried, mainly unsuccessfully.144 Also of concern was the limited understanding of the health impacts of smoking. Most were aware of the risks of cancer but had little other knowledge about the risk of smoking in pregnancy, the impact of exposure to environmental tobacco smoke or nicotine dependence and its treatment.

84 PART 4: THE HEALTH IMPACT OF TOBACCO The health impacts of tobacco are daunting. Tobacco has been recognised as the single largest avoidable cause of premature death and the most important known human carcinogen.69 Half of all long-term smokers will eventually be killed by tobacco and of these, half will die during middle age, losing 20-25 years of life. 145 ,146 ,147 Tobacco currently kills 4 million adults annually world wide, accounting for one in 10 adult deaths. By 2030 or sooner, these figures will reach 10 million and one in six deaths.145 The number of DALYs lost to tobacco use is projected to rise from under 40 million years in 1990 to over 120 million years in 2020.148

4.1 Chronic disease

The adverse impact of tobacco on health was first reported over 200 years ago in relation to carcinoma of the lip87, 149 and was subsequently confirmed through a growing series of case control87,150,151,152,153,154 and cohort studies,155,156,157,158,159,160 most notably the two large Cancer Prevention Studies (CPS I and II) which followed separate cohorts of over 1 million citizens from 1959 and 1982161, 162 and the UK Doctors study which followed 34,000 doctors over 40 years from 1951.146

Smoking has now been positively associated with over 40 diseases and negatively associated with 8 or 9 more and the list of links contintues to grow.160 For most diseases the evidence is strong and the associations between smoking and mortality are chiefly causal in character: findings have been confirmed in numerous studies in different populations, biological mechanisms are understood, the association is strong and a dose response relationship is seen. The diseases fall into three main categories, cancer, vascular diseases and chronic lung diseases (Table 1-20).

Worldwide, tobacco is the main known cause of human death from cancer.163 Lung cancer alone has the highest incidence and mortality of any cancer in men, whilst in women it is the fourth most common cause of cancer but the second most common cause of cancer death. Moreover, in eastern Europe the incidence of lung cancer in men is higher than in any other world region.163

85 Table 1-20 Relative risks in current versus never male smokers (except for cervical cancer in female smokers) based on UK Doctors study and CPS II. Adapted from data in Doll160, Wald164 and Boyle149 Increased risk largely or entirely due to smoking Increased risk partly due to smoking Increased risk largely or partly due to confounding Cancers: Mouth, pharynx, larynx* 24, 24.5 Myeloid leukaemia $(sig relation with amount Liver (causal and confounding) 1.6 Lung* 15, 22.4 (THUN: 23.2) smoked in UK Doctors study) Cervix (largely confounding) --,2.1 (CPS data only) Oesophagus* 7.5, 7.6 Nose and nasopharynx $ Large bowel (largely confounding) Bladder* 2.3, 2.9 Kidney $ 2.1, 3 Pancreas* 2.2, 2.1 Stomach $ 1.7

Cardiovascular diseae: Ischaemic heart disease 1.6, 1.9 Hypertension 1.4, 2.4 Myocardial degeneration 2.0 Pulmonary heart disease - Aortic aneurysm 4.1, 4.1 Peripheral vascular disease 2.0 Cerebral vascular disease 1.3, 2.2 Respiratory diseases: Pneumonia 1.9 Chronic obstructive pulmonary disease 12.7, 9.7 (THUN 11.7) Asthma 2.2, 1.3 Other diseases: Gastric ulcer 3.4 Cirrhosis 5.3 Duodenal ulcer 4.1 Suicide 1.6 Crohns disease 2.1 Poisoning 2.7 Osteoporosis Periodontitis 3.0 Tobacco amblyobia Age-related macular degeneration Cataracts 2.2 Hip fracture 1.3 * defined as causal by IARC 69, $ considered causal later – see Doll165

86 4.2 Dose response

The risks posed by tobacco vary with the amount smoked and duration of smoking. Dose response relationships have been measured in a number of ways: the number of cigarettes smoked per day, age of first smoking and total number of years smoked, degree of inhalation, number of puffs per cigarette, tar and nicotine levels and various combinations of these variables.166 For most causes of death related to tobacco, risks increase progressively with the amount smoked and decline over time in ex-smokers to levels between current and non-smokers.146,161

The most detailed information is available for lung cancer; this shows that the risk is dependant both on the duration of smoking and the number of cigarettes smoked daily, the duration being more important.167,168 Thus the earlier the age at initiation of smoking, the greater is the risk. A recent study found that those who started smoking before the age of 15 compared with those who had started aged 20 or older had twice the risk of lung cancer.169 Similarly, the longer the time period during which a major proportion of adults in a population have smoked, the greater the incidence and mortality from lung cancer in that population. For this reason, risks of lung cancer have been found to be progressively higher in studies conducted later in the tobacco epidemic.169,161 The impact of tar and nicotine yields is somewhat more complex with a recent study confirming findings of earlier studies: Smokers of high tar (>22mg), non-filter cigarettes have a significantly greater risk of lung cancer than smokers of medium tar (15-21mg), filter cigarettes whose risk does not differ from smokers of low (8-14mg) or very low tar (<8mg) cigarettes.170 Detailed work on ischaemic heart disease (IHD) mortality shows that risks also increase with the amount smoked but that the relative risk of smoking is highest in younger age groups, although due to the underlying patterns of death, absolute risks of death from IHD are higher in older age groups.171

4.3 Delay in health impact

At a population level there is a three to four decade delay between the rise in smoking prevalence and its health impact. Smoking attributable mortality is therefore most closely related to smoking patterns 30 or more years previously, rather than to current prevalence. However, the delay in the onset of smoking-related disease varies according

87 to the particular disease. With lung cancer, health effects are not generally apparent until 20-30 years after smoking becomes widespread in a population and do not reach their peak until 30-40 years after the peak in smoking prevalence.172 For Chronic Obstructive Pulmonary Disease, where the mechanism of organ damage due to smoking is similar to that of lung cancer, the time delay may be similar. The effect on vascular diseases may occur somewhat earlier, suggested by the fact that the excess mortality from IHD in smokers is greater in middle than older age and by the observation that quitting is associated with a more rapid decline in IHD mortality than is seen with lung diseases.173

4.4 The tobacco epidemic

A four stage model of the smoking epidemic has been described, based on observations of trends in cigarette consumption and tobacco related diseases in western countries with the longest history of cigarette use, namely the UK and the US.172

The model describes the rise in male smoking, followed by the rise in female smoking, a plateauing and then fall in smoking prevalence and decades later the rise in tobacco related mortality. It is illustrated graphically in Figure 1-3 and detailed in Table 1-21.

In addition to the delay between the rise in smoking prevalence and its health impact, a number of other trends can be noted in the model - the one to two decade delay between the increase in male and female smoking and the changing social pattern of smoking. Smoking is initially more common amongst the upper classes, but as the better educated quit, this pattern reverses.

88 Figure 1-3 The four stage evolution of the tobacco epidemic

Source: Lopez et al 172

89 Table 1-21 The four stages of the tobacco epidemic MALE FEMALE Cigarette Tobacco control Smoking prevalence Deaths from tobacco Smoking prevalence Deaths from tobacco consumption

Stage 1 Comparatively low At the start, deaths from Rare 0-10% As in non-smoking Per capita Tobacco control (1-2 <15% but increasing tobacco similar to non- populations consumption low underdeveloped. decades) smoking populations. By the <500/adult (most Smoking becomes end of this phase some male consumed by men) socially acceptable deaths attributable to tobacco. Stage 2 Continues to increase By the end of this phase Increases rapidly, but Rise in lung cancer rates Per capita Tobacco control not well (2-3 rapidly, peaks at 50- ~10% male deaths due to lags behind men by 1-2 much less, may reach 8-10 consumption 1000- developed. decades) 80% tobacco. decades. per 100,000 3000 (majority Relatively few ex- Male lung cancer rates consumed by men) smokers. increase approximately 10 Men 2000-4000 Smoking rates similar fold from 5 to 50 per in different social 100,000. classes and may be higher in upper classes. Stage 3 Plateaus at a high Rapid rise in smoking- Plateaus at a lower Still relatively low (5%) but Men 3000-4000 Tobacco control (3 level, then starts to related mortality (to level than in men (max rising. Women: 1000-2000 improved. decades) decline. approximately 30% deaths) 35-45%) and often for By the end of this phase lung Smoke free public places Lower in middle and and higher in middle age. prolonged period, then cancer rates reach and transport (but not older age men who By the end of this phase lung declines late in this approximately 23-40 per workplaces) become give up. cancer rates peak at around phase. 100,000. standard. Male and female 110-120/100,000 Marked age gradient Smoking becomes less prevalence rates with higher rates in socially acceptable. become more similar. young women. Declines in prevalence Declines in prevalence greater among better greater among better educated. educated. Stage 4 Continues to decline Peak in smoking-related Continues to decline Rising rapidly. Smoke free environments but slowly. Male mortality at 30-35% of all but slowly. Will peak at ~20-25% of all become an issue. prevalence might be deaths (40-45% in middle Social differences deaths 2-3 decades into this around 33-55%. age) during the early part of persist. phase (approximately 2 Social class this stage. decades after the male peak) differences persist. About 1 decade later, and decline thereafter. mortality from tobacco will start to decline.

90 4.5 Impacts of tobacco in non-western populations

Most of the evidence on the health impacts of tobacco use comes from high-income countries. More recently, however, evidence has become available from China and India.174,175,176,177 These studies suggest that, although the overall risks of smoking are about as great as in high income countries and the diseases caused by smoking are similar, the specific patterns of smoking-related diseases may differ.

For example, in China, the relative risk of death from all causes in smokers versus non- smokers was almost two-fold, as reported in the west,175 but the relative risk of lung cancer in Chinese males was approximately 3,145,175,176 compared with 20 in the US. The lower relative risk is likely to reflect both the shorter duration of cigarette smoking, the relatively late age of first smoking and perhaps also the relatively high background rate of lung cancer not related to smoking. China is still early in its epidemic and hence the risks from tobacco have yet to be realised in full.

The absolute or attributable risks of tobacco in China also differ from those seen in the west, reflecting the different underlying causes of death. Deaths from IHD contribute a much smaller proportion of the total number of deaths caused by tobacco whilst respiratory diseases and cancers account for most (about 60%) deaths. Smoking kills by making diseases that are already common somewhat more so.

There has only been one analysis of cohort study data in the former Soviet Union that can be used to assess the risk of tobacco use there, using data from the Lipid Research Clinic study, and published in Russian (Figure 1-4).178 This study, based on a sample from Saint Petersburg, Russia, suggests that the relationship between smoking and mortality is similar in middle-aged men there and in the west.

91 Figure 1-4 Survival by smoking status among men enrolled in the Russian arm of the Lipid Research Clinics study

1.0

0.9

0.8

0.7

0.6 Never smoked 0.5 Ex-smokers

0.4 Smokers: low

Survival rate . Smokers: medium 0.3 Smokers: heavy 0.2

0.1

0.0 40 45 50 55 60 65 70 75 80 Age

Source: Deev et al, 2000

Despite these similarities, care must be taken in assuming that findings in one population apply to other population groups. The relative risks of disease from tobacco will depend on a number of factors. First, the presence and prevalence of other aetiological agents such as alcohol, diet, blood lipids or other cardiovascular risk factors and exposure to asbestos or infective factors with which smoking may interact. For example there is convincing evidence that a diet rich in vegetables and fruits exerts a protective effective against lung cancer and that alcohol and tobacco consumption act multiplicatively to increase the risk of certain cancers arising from each factor alone.163 A second issue is the duration of smoking, or at a population level, current as well as previous smoking patterns (the stage of the epidemic the country has reached). Related to this, the impact will also depend on the type of cigarettes smoked, the tar yields, the age of first smoking and the number smoked per day. The absolute or attributable risks of illness due to smoking will also vary from country to country for similar reasons and in particular as a result of variations in underlying causes of illness.

92 As we will see below, the FSU has very high rates of tobacco related disease. Potential explanations include the high tar yields of traditional filterless cigarettes, a diet low in anti-oxidants and high alcohol consumption. Confounding between alcohol and tobacco is important; those that smoke also tend to drink. As alcohol is now thought (in western populations) to be cardio-protective, this confounding may cause the relative risk of cigarette smoking on vascular diseases to have been underestimated in published studies. In the FSU, however, drinking patterns are different, and it is likely that alcohol increases the risk of IHD179 and could potentially interact differently with smoking.

4.6 Health impacts of tobacco in the FSU

To help overcome the lack of accurate data on the health impacts of tobacco and on smoking histories in much of the world, Peto et al have attempted to model the number and proportions of deaths due to tobacco.145,147 They assess the absolute excess mortality from lung cancer in each country by comparing national lung cancer mortality rates with rates in US non-smokers and use this as an indication of the population’s exposure to tobacco. From this, they calculate the proportions of deaths from other causes that can be attributed to tobacco based on the American Cancer Society study. o These data provide the first ever such estimates for the fifteen NIS, although there are uncertainties about the quality of population and mortality data particularly for some of the central Asian republics and the applicability of the risk estimations to these countries.

A number of important findings can be noted. First, the data confirm the considerably greater risk of death in middle aged men in the FSU and other former socialist states than other developed countries, with the risk of death more closely approximating that in India or sub-Saharan Africa.

Second, the death toll from tobacco amongst males is extremely high in the FSU and its former socialist neighbours, higher than elsewhere in the developed world. For example, approximately 20% of men aged 35 will, on the basis of 1990 mortality levels, be killed by tobacco before age 70 in the FSU compared with 10% in the EU (Figure 1-5, Table

o To be conservative no non-medical deaths (eg from fires), no neonatal deaths and no deaths before the age of 35 are attributed to tobacco. In addition, the calculated excess percentage of deaths from other causes attributed to tobacco is halved, thereby preventing overestimation of tobacco’s impacts (at the risk of potentially underestimating its impacts).

93 1-22).145 Among middle aged men in the FSU, 56% of cancer deaths and 40% of all deaths are attributed to tobacco compared with 47% and 34% in the EU (Table 1-22). Indeed the authors conclude that the effects of tobacco are so great amongst men in the former socialist countries that the proportion of deaths attributed to non-communicable disease is higher in this region than in any other world region, that tobacco use explains the greater part of the excess risk of death in middle age in former socialist compared with other developed countries, and accounts (in men) for up to 50% of the excess mortality between East and West Europe.44,145 The exact contribution of tobacco to mortality, measured in this way, varies between the countries of the FSU, with the greatest impacts seen in Russia and the other European countries and the lowest impacts in central Asia. Once again Kazakhstan is an exception, with the risks from tobacco more similar to the European countries of the region, whilst in Moldova the risks are slightly lower than in the other European countries (Table 1-22).

Third, among women the death rate is far lower than in men, making the estimates of tobacco’s impacts somewhat less reliable. Nevertheless, the risks are found to be considerably lower than in countries such as the US or UK where the tobacco epidemic in women is further developed, but similar overall to risks in all EU or developed countries combined (Figure 1-5, Table 1-23). The individual country data, as in men, suggest tobacco makes a higher contribution to mortality in the European countries and Kazakhstan and a lower contribution elsewhere in central Asia. (Table 1-23).

Fourth, as a result of the high rates in men and the low rates in women, the gender gap in tobacco related-disease is far wider in the former socialist countries than other regions studied. And, unlike in men, tobacco use cannot explain the somewhat higher risk of death in middle aged women in the former socialist compared to other developed countries.

Finally, trends over time from 1955 to 1995 show that, among men, smoking-related diseases are increasing rapidly in the FSU and other socialist countries whilst stabilising in the EU and US and declining in the UK (Figure 1-6). As a result, the FSU countries are projected to have the highest risks in the world of male tobacco-attributable death in middle age. There is some concern about the accuracy of the earliest data in this period,

94 but trends from 1975 to 1990 are similar, with tobacco-related mortality in middle aged males in former socialist countries increasing, whilst elsewhere (apart from Portugal) this is constant or decreasing. Moreover, the fact that total mortality is increasing in the former socialist economies p while mortality not attributable to tobacco is steady, suggests that the increase in total mortality is almost entirely attributable to tobacco. Similarly, when looking only at cancer deaths, the data suggest that tobacco alone can account for almost all the between country differences in adult male cancer mortality and almost all the change in mortality.

In contrast, tobacco related mortality in women is increasing in almost all developed countries, with rates increasing more slowly in the FSU than in the EU or US whilst declines are again seen in the UK (Figure 1-6). These trends reflect the differing stages of the tobacco epidemic in these various countries and regions and suggest that amongst women in the FSU, although risks from tobacco are now low, substantial hazards are likely to emerge.

This evidence is supported by a geographically more extensive analysis, based on the same methodology, which showed that the accumulated risk of smoking in men under 70 years of age was higher in the FSU and Eastern Europe than any other world region.180 Above this age the burden was higher in western Europe and the US, an indication of their longer history of smoking. Trends with age showed a sharp rise in the health impact of smoking in younger generations in the FSU, particularly in its European countries, compared to more steady trends in Western Europe and the US, again suggesting the epidemic is more recent. This again implies that as this younger generation ages, the health impacts of tobacco will continue to increase. In women, the health impacts in different age groups were steady in the non-European FSU countries, but showed a slight increase in younger age groups in the European countries.

p It should be noted that all other developed countries except Denmark experienced an improvement in mortality over the period 1975 to 1990.

95 Figure 1-5 Probability of dying from tobacco in middle age (age 35-69) in the FSU and the EU, based on 1990 data

25

20

15 Male Female 10

5

0 FSU Former Socialist OECD developed EU (12)

Source: Peto et al147

Figure 1-6 Proportion of all deaths attributed to smoking in men and women aged 35-69,1955-1995

60

50

Former Socialist economies Male Former Socialist economies Female 40 FSU Male FSU Female EU (12) Male 30 EU (12) Female UK Male UK Female 20 US Male US Female

10

0 1955 1965 1975 1985 1995*

*1995 data are estimates. Source: Peto et al147

96 Table 1-22 The Risks of dying and proportion of deaths attributed to tobacco in men in 1990 In men aged 35-69 years In men of all ages Probability of % Cancer % cancer 35 year old deaths % Total deaths deaths % Total deaths dying before attributed to attributed to attributed to attributed to Country/Region age 70 smoking smoking smoking smoking Armenia1552384523 Azerbaijan1035242914 Belarus1853394626 Estonia1955384826 Georgia 9 41 24 36 15 Kazakhstan 22 59 43 52 28 Kyrgyzstan 13 42 28 36 17 Latvia 20 55 38 47 25 Lithuania1855384625 Moldova1549314220 Russia 22 59 42 52 30 Tajikistan 5 21 14 17 5 Turkmenistan 10 38 22 30 9 Ukraine2056405028 Uzbekistan83420268 All developed1349354224 OECD developed1045323923 Former Socialist1956394826 FSU2056405027 EU (12)*1047344024 *the then 12 members of the EU Source: Peto et al147

97 Table 1-23 The risks of dying and proportion of deaths attributed to tobacco in women in 1990 In women aged 35-69 In women all ages

Probability of 35 % Cancer deaths % Total deaths % cancer deaths % Total deaths year old dying attributed to attributed to attributed to attributed to Country/Region before age 70 smoking smoking smoking smoking Armenia 2 5 643 Azerbaijan 0 0 000 Belarus <1 1 2 0.9 0.7 Estonia24643 Georgia<12210.6 Kazakhstan31112107 Kyrgyzstan 1 3 4 4 4 Latvia25643 Lithuania <1 3 3 3 3 Moldova12333 Russia25754 Tajikistan00000 Turkmenistan00000 Ukraine25654 Uzbekistan 1 5 5 4 2

All developed 2 11 12 10 7 OECD developed 3 14 16 12 9 Former Socialist 2 6 7 5 4 FSU25654 EU (12)*18975 *the then 12 members of the EU Source: Peto et al147

Table 1-24 Standardised death rates from various causes in countries of the CIS, 1999 data* SDR, trachea/bronchus/lung SDR, ischaemic heart disease, 0- cancer, all ages per 100000 64 per 100000 Male Female Male Female Armenia 53.72 6.35 122.88 29.72 Azerbaijan 30.23 5.52 177.04 67.11 Belarus 82.12 5.33 218.64 59.41 Georgia 33.75 5.36 139.27 46.45 Kazakhstan 80.15 11.96 193.83 58.68 Kyrgyzstan 32.37 5.3 119.09 41.05 Republic of Moldova 46.14 8.85 138.58 61.58 Russian Federation 88.68 8.74 208.15 50.12 Tajikistan 9.29 3.16 86.84 44.24 Turkmenistan 17.51 4.38 174.57 67.84 Ukraine 72.27 8.35 197.72 62.95 Uzbekistan 16.33 4.65 121.07 57.56 EU average 66.63 15.86 38.18 8.85 CSEC average 85.26 15.13 90.16 23.94 CIS average 71.37 7.94 188.38 54.19 Source: WHO Health For All Database updated June 1993. http://hfadb.who.dk/hfa/ (last accessed 8/12/03) *1999 data except for Turkmenistan where 1998 data used

98 In addition to these indirect estimations of tobacco’s impacts, examination of specific causes of smoking-related diseases and death provides further insights into the health impacts of tobacco (Figure 1-7-Figure 1-9). For example, the FSU exhibits extremely high death rates from IHD (Figure 1-8). In 2000, rates of IHD among men were four times higher in the CIS than in the EU and almost five times as high in men aged under 65. Among women the gaps are even greater, with rates in the CIS 4.6 and 6 times higher respectively. These figures also show clearly how mortality from IHD changed rapidly at the time of transition – climbing from 1991 to 1994, falling from 1994 to 1998, and then rising once again. In both genders these fluctuations are more marked in the under 65s, who, as described above, were most influenced by the stressful impacts of transition. Although smoking is an important risk factor for IHD, it cannot explain these rapid variations in or the high rates of IHD among women.

The traditional risk factors for IHD identified in western epidemiological research, (age, sex, family history, blood pressure, cholesterol and smoking), must play a role in the FSU.181 But as in other countries, there is growing evidence that the predictive power of these risk factors is poorer than had been assumed when applied to populations other than the often atypical ones from which they were derived. 182 , 183 Thus it is also becoming clear that these fail to explain the high cardiovascular mortality rates in the FSU.184,185 The much higher rate of sudden cardiac death among young men is also unexplained.43 There is, however, growing evidence that the low level of dietary antioxidants, due to the low intake of fruit and vegetables, is likely to play a role186,43,184 and dietary changes have been suggested as a cause for the rapid decline in cardiovascular mortality seen in Poland187 and the Czech Republic post transition.188 Alcohol is also thought to be important.3,184,43 In all of northern Europe, but especially in Russia and its neighbours, alcohol is typically drunk as vodka and in binges,189 unlike the more steady consumption in southern and western Europe. An increasing volume of research has examined the possible involvement of alcohol in cardiovascular mortality in eastern Europe50, 190 , 191 and suggests that episodic heavy drinking is consistently associated with an increase in the risk of cardiovascular death, especially sudden cardiac death.179 As suggested above, psychosocial stress and lack of control over life events may also play a role.44

99 Nevertheless, national patterns appear to broadly reflect both the general mortality patterns seen in the region (as described above) and patterns of smoking, with the highest rates of IHD in men under 65 observed in Russia, Ukraine and Kazakhstan, the lowest rates in central Asia (other than Kazakhstan), with other countries some way between (Figure 1-8). In women the pattern is somewhat different, with low rates again seen in Kyrgyzstan, Tajikistan and Armenia but high rates in Azerbaijan, Ukraine and Moldova, for reasons that remain inadequately understood.

Lung cancer data are more easily interpretable, because in developed countries, smoking, especially when of long duration and heavy intensity, accounts for approximately 90% of lung cancer in men. 163 The proportion is somewhat less in women, largely because their as yet lower absolute levels mean that non-smoking related cancers, such as alveolar cell carcinoma, account for a greater share of the total. Between country comparisons are also more likely to be valid - lung cancer is almost always fatal, health care has little impact on survival and variations in disease prevalence due to other underlying causes or differences in coding of deaths should be less with lung cancer than for other smoking-related diseases.

In 1932 compulsory cancer registration was introduced in the Ukraine and in 1953 extended to the whole of the USSR. The International Classification of Diseases (ICD) system has been used for cancer deaths since 1965, although in most ex-Soviet countries it only replaced a more basic Soviet classification system used to report general mortality data in the mid 1990s. This system mapped to the ICD; the difference was that as it had fewer individual codes it lacked the precision of ICD. However, for the purposes of monitoring deaths from cancers at major sites, the results were the same.192 From 1965, therefore, it has been possible to track trends in cancer incidence. These show a steady rise in lung cancer incidence, particularly in men, so that by 1980 (and somewhat sooner in a few of the republics) it became the most frequent tumour in males.192,193 Age cohort analysis has shown how the increase began with cohorts born around the turn of the century. Subsequent data show that incidence rates have continued to rise, most notably in the Baltic republics, although they are now levelling off (Figure 1-7). Trends in the other European ex-Soviet countries are broadly similar. There are however marked differences in incidence rates between the FSU countries,

100 with the Baltic States, Russia, Ukraine and Belarus having considerably higher rates than elsewhere.

In women, incidence rates are considerably lower and few clear trends can be observed other than in Estonia, where lung cancer incidence is increasing rapidly. This is consistent with the data on smoking rates described above, suggesting that women in Estonia have been smoking for longer than those elsewhere in the region and implying that lung cancer will soon increase in frequency elsewhere, in line with the rest of Europe.

Given its high fatality rate, lung cancer mortality rates are similar to incidence rates, with male mortality rates rising until the early- to mid-1990s (Figure 1-9).194 Although a decline in lung cancer mortality in the CIS is then observed, this is thought to be a cohort effect with those who were teenagers between 1945 and 1953 carrying forward lower smoking rates as cigarettes, like other consumer goods, were in short supply in the period of post-war austerity under Stalin.180,195 The presence of cohort effects can also be inferred from inspection of lung cancer death rates, which among men over 65 are only slightly higher in the CIS than in the EU while among those under 65 the rates in Russia are almost twice as high as in the EU. Consistent with the findings detailed above,180 this reflects how male smoking rates have been declining across most EU countries for the last 25 years, so that peak rates in male lung cancer mortality were reached in northern Europe in the late 1970s and in southern Europe in the late 1980s.194 In contrast, male smoking rates in the CIS have not yet declined and thus cancer rates in the younger generation remain higher than in the EU.

Among men there are wide variations in national lung cancer mortality rates amongst the countries of the CIS, with the highest rates seen in Russia, Belarus, Kazakhstan and Ukraine and the lowest in the other central Asian republics, the Caucasus and Moldova (Figure 1-9). The rates in the first four countries are amongst the highest in the world – with Russia having the 5th highest lung cancer mortality rate in the world (based on 2000 data) after Hungary, Poland, Belgium and Croatia.196 Meanwhile the rates in the other central Asian republics, the Caucasus and Moldova are amongst the very lowest in Europe.117

101 Among women, lung cancer rates are far lower than in men and considerably lower than in the EU, reflecting the historically lower smoking rates. An increasing death rate in women in the EU and central and south-eastern European countries is expected given that smoking rates in women peaked later than in men and although now declining in some EU countries, continue to rise elsewhere, for example in much of southern Europe.117,197 Once again the decline in the CIS is unexpected, although the recent increases in female smoking rates would not yet have impacted on lung cancer rates and it is therefore possible that the decline in mortality among women is due to the same cohort effect observed in men.194,195 Unsurprisingly, between country differences are less marked than in men, with death rates from lung cancer in the central Asian republics other than Kazakhstan amongst the 5 lowest in Europe and 11 of the NIS having amongst the 15 lowest rates in the 47 European countries for which data are available117(Figure 1-9).

102 Figure 1-7 Incidence of lung cancer per 100,000 1981-2001 for all countries of the FSU

Males of all ages Armenia

Azerbaijan 120 Belarus Estonia 100 Georgia

Kazakhstan 80 Kyrgyzstan 60 Latvia Lithuania 40 Republic of Moldova

Russian Federation 20 Tajikistan

0 Turkmenistan Ukraine 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 Uzbekistan

Females of all ages Armenia Azerbaijan 25 Belarus

Estonia 20 Georgia

Kazakhstan

15 Kyrgyzstan Latvia

10 Lithuania Republic of Moldova

Russian Federation 5 Tajikistan Turkmenistan 0 Ukraine 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 Uzbekistan

Source: WHO Health For All Database updated June 1993. http://hfadb.who.dk/hfa/ (last accessed 8/12/03)

103 Figure 1-8 Standardised death rates per 100,000 from ischaemic heart disease, 1981 - 2001 Males all ages Females all ages 700 400

600 350 300 500 EU 250 average 400 200 CSEC 300 average 150 200 CIS 100 average 100 50 0 0

1 3 5 7 9 1 3 5 7 9 1 8 8 8 8 8 9 9 9 9 9 0 81 83 91 93 9 9 9 9 9 9 9 9 9 9 0 995 1 1 1 1 1 1 1 1 1 1 2 19 19 1985 1987 1989 19 19 1 1997 1999 2001

Males250 aged 0-64 Females aged 0-64 70 60 200 50 150 EU 40 average

CSEC 100 30 average 20 CIS 50 10 average

0 0

3 3 8 7 9 981 9 8 3 1981 1983 995 1 1 985 9 7 1 1985 1987 1989 1991 199 1 1997 999 1 1 198 991 995 9 0 1 2001 1 199 1 9 0 1 1999 2

104 Figure 1-8 Standardised death rates per 100,000 from ischaemic heart disease, 1981 - 2001 (continued) Males all ages by country Males aged 0-64 by country 800 300

700 250 600 Belarus 200 500 Georgia Kazakhstan 400 150 Republic of M oldova

Russian Federation 300 100 Ukraine 200 Uzbekistan 50 100

0 0

3 5 1 8 9 1 87 9 1 98 9 5 7 1 198 985 9 8 7 9 1981 1 19 1987 98 9 9 1983 1 1 9 9 9 1 1 19 1993 00 1 199 1993 1995 9 9 19 19 1999 2 1 1 200

Females all ages by country Females aged 0-64 by country

600 90 500 80 Belarus 70 Georgia 400 60 Kazakhstan

Republic of M oldova 300 50 40 Russian Federation 200 30 Ukraine 20 Uzbekistan 10 0 10 0 0 1999 2001 2001 1997 1997 1999 1993 1995 1993 1995 83 1991 105 1989 1991 1987 1989 1985 1987 981 1985 1981 1983 1 19 Figure 1-9 Standardised death rates per 100,000 from cancer of the trachea, bronchus and lung, 1981-2001 Males all ages Females all ages

18 100 90 16 80 14

70 12 EU 60 10 average 50 8 CSEC 40 average 6 CIS 30 average 20 4 10 2 0 0 1 1 7 0 1 7 9 95 9 0 8 85 8 9 9 1999 2 9 9 1989 1 1993 1 19 81 9 1 3 1 1983 1 19 83 9 1 19 19 1985 1987 198 199 199 1995 1997 199 200 Males aged 0-64 Females aged 0-64 60 10 9 50 8 40 7 6

30 EU 5 average 4 CSEC 20 3 average CIS 2 average 10 1 0 0

9 5 1981 87 8 91 8 7 9 1 1983 1985 19 9 9 8 8 7 9 1 1 99 01 1981 1983 19 19 9 9 1 1993 1995 1997 9 0 19 199 1993 0 1 2 1995 19 19 20

106 Figure 1-9 Standardised death rates per 100,000 from cancer of the trachea, bronchus and lung, 1981-2001 (continued) Males all ages by country Males aged 0-64 by country 120 70

100 60

Belarus 80 50 Georgia

60 40 Kazakhstan Republic of Moldova 30 40 Russian Federation 20 Ukraine 20 Uzbekistan 10

0 0

1 8 3 7 9 9 8 8 8 5 1 83 9 1 9 9 9 93 9 0 981 85 1 1 1985 1 1 1991 9 9 0 1 19 993 97 99 1 1 1997 1999 2 19 1987 198 1991 1 995 Females all ages by country Females 0-64 by country 1 19 19 200 8 16 7 14 6 Belarus 12 Georgia 5 10 Kazakhstan 4 Republic of M oldova 8 Russian Federation 6 3 Ukraine 2 4 Uzbekistan 2 1

0 0

7 9 9 99 01 1 87 8 93 95 9 0 8 85 9 1 2 9 19 1991 1 19 19 993 995 997 999 01 19 1983 1 19 1981 1983 1985 1987 1989 1991 1 1 1 1 20 Source for all the above graphs: WHO Health for All Database http://hfadb.who.dk/hfa/ last accessed 8/12/03

107 4.7 Gender differences in mortality The gender gap in mortality in the FSU is the widest recorded anywhere in the world and has grown over recent years, mirroring the mortality trends during the transition (increasing until 1994, then falling and rising again post 1998) (Figure 1-10).23

Research undertaken early in the 1990s suggested that over 60% of the mortality gap was due to two causes of death – injuries and poisoning and cardiovascular disease. The widening of the gap is due mainly to the increasing differences in mortality from cardiovascular diseases and cancer, as well as from injury.45 More recent work highlights the important contribution of tobacco to the gender gap in Europe. It suggests that once the impacts of tobacco are removed the relative gender gap (the mortality rate ratio) between east and west disappears,198 a finding consistent with the reported causes of death that drive the gender gap.

Figure 1-10 Trends in the gender gap in life expectancy at birth in the European Union and the Newly Independent States, 1990-2001

14

12

10

8 EU average NIS average 6

4

2

0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Source: WHO Health for All Database http://hfadb.who.dk/hfa/ last accessed 8/12/03

108 4.8 Lifestyles and health promotion in the USSR/FSU Soviet lifestyles were, and remain, generally unhealthy, characterised by smoking, excess alcohol consumption, high fat diets and lack of exercise, particularly in middle aged, working-class males.199 Although the USSR made good progress in reducing mortality from common infectious diseases, in particular where they were preventable by immunisation, the Soviet system was much less effective at responding to the complexity of non-communicable diseases, such as cardiovascular and metabolic disorders. 200 The Soviet authorities were generally slow to act on evidence about disease causation or apply scientific knowledge.201 The only exception is perhaps the 1980s anti-alcohol campaign, although this was undertaken primarily because of the impact of heavy drinking on industrial production.23

A useful comparison of health promotion responses can be made between Finland and neighbouring Estonia. In the 1960s mortality from chronic diseases began to increase in the west as it did in the FSU.201 The Finnish authorities responded with a major health promotion initiative, the North Karelia project, which was formally evaluated, shown to have produced a major decline in cardiovascular mortality, and rolled out through the rest of the country.202 In contrast, there was no systematic response in the Soviet Union, nor in newly independent Estonia until collaboration with Finland helped initiate appropriate responses including the establishment of health information systems and health promotion interventions.202

As well as health system failures, it is suggested that through the social and psychological conditions it engendered, socialism constrained the practice of positive health lifestyles. The collectivist viewpoint reduced individual responsibility for health and encouraged passivity and irresponsibility. If sickness occurred, the state would provide.199 A low sense of control over one’s life has been observed in the region and shown elsewhere to relate to health behaviours,60,57 although low control was not a determinant of smoking in my previous work in Ukraine.129 The ability to choose positive health behaviours was constrained and the policy orientation was to invest responsibility for health in the state, not the individual. Social life, especially among men, centred on negative health practices, especially drinking and smoking. Vodka and

109 cigarettes were some of the few consumer goods that the system could provide, and so became a necessary mechanism to re-circulate currency.

Although the communism has now collapsed, these passive attitudes and unhealthy lifestyles have continued, or even worsened, but as the Soviet safety nets have disappeared, the dangers have increased. Recent commentators have described the Moscow lifestyle as “passive” and “unhealthy” with healthy lifestyle choices limited by material constraints.199,203 Alcohol and cigarette consumption and the proportion of fat in the diet are amongst the highest in the world.199 Moreover, there is compelling evidence that alcohol consumption increased markedly following transition. The government ended its monopoly on vodka production, allowing unrestricted sales from other sources and costs dropped significantly. Thus while, when defined in terms of a standard basket of consumer goods, the purchasing power of the average wage fell by nearly a half in 1992-3, the purchasing power relative to vodka increased three times. It has been argued that the lower alcohol costs were a conscious attempt by the Yeltsin government to calm discontent in the face of the country’s economic woes by making alcohol more affordable,199 reminiscent of Catherine the Great’s observation that a drunken population was more compliant.

Experts working in the former socialist countries have identified a number of important challenges and barriers to effective tobacco control in the region. These include the enormous economic challenges and competing priorities for scarce resources, such that health promotion struggles to be prioritised; lack of cooperation and communication between the ministry of health and other ministries (whose priorities may well be focused on attracting investments from TTCs and other companies); the lack of data on smoking behaviour and its health and economic impacts with which to press for policy changes; poor customs and other controls on smuggling; and the absence of well developed non-governmental organisations (NGOs).85, 204 It is also suggested that privatisation of the tobacco industry and the presence of the TTCs may pose a greater threat to tobacco control than the previous state owned companies.204 In addition to tobacco industry resistance to legislation it has been suggested that the population, after years of totalitarian control, may also resist tobacco control measures, particularly in an

110 environment where alcohol and narcotic misuse are perceived as far more serious problems. 204

111 SECTION C: AIMS AND OBJECTIVES 5.1 Gaps in knowledge

The literature review presented above outlines the profound changes that occurred in the FSU during transition including momentous changes to the region’s tobacco industry. It details evidence from the TTC’s previous entry to Latin America and Asia, which indicates that TTC entry led to increases in tobacco consumption and efforts to undermine tobacco control. This evidence raises concerns about the potential impact of TTC entry to the FSU, a region which already has the worst mortality profile from tobacco in the developed world, and very poorly developed health promotion and public health systems. However, given the different circumstances and methods of entry, particularly from those seen in Asia, direct analogies are dangerous.

A number of gaps in current knowledge can therefore be elucidated:

• Despite the major changes to the FSU’s tobacco industry, there has been no systematic research into its nature, scale or impacts.

• Although anecdotal reports of the TTC antics, for example, in exploiting or flaunting tobacco control legislation abound, suggesting the TTCs have undermined tobacco control policies in the FSU, there has been no systematic research into these allegations.

• As they have never been systematically studied, little if anything is known about the impact of tobacco industry privatisation and investment liberalisation on tobacco trade or consumption and hence public health, yet the FSU experience provides a unique opportunity to study this issue.

• There are no reliable published analyses of trends in tobacco or cigarette consumption in the FSU available beyond the early 1990s.

• There are very few accurate, recent data on smoking prevalence in the FSU, particularly in the Caucasian and central Asian republics.

Closing these knowledge gaps is important for a number of reasons. First, at the time of transition tobacco industry privatisation was promoted by the IFOs, who treated privatisation of this industry like that of any other, seemingly oblivious to the fact that

112 tobacco is uniquely damaging to health. Tobacco industry privatisation continues in other parts of the world, in some instances still as a result of pressure from the IMF. There is therefore a pressing need to understand the potential impacts of tobacco industry privatisation and how any negative impacts may be countered. Second, country specific data on smoking behaviour, (particularly by age, sex and area), are widely recognised as pre-requisites for designing appropriate prevention policies.11,12,13 and are generally considered useful in pressing for policy changes.85,204 The lack of such data for much of the FSU therefore represents a serious impediment to tobacco control. Moreover, to date, other than across the Baltic States, no countries of the FSU have collected data in comparative format, thus precluding accurate comparison of smoking behaviour and determinants. Third, although the TTCs previous entry to other regions has been described in detail, with authors attempting to outline the strategies adopted by the TTCs to penetrate new markets in Asia and Latin America, such efforts have never been studied from the industry’s perspective as the industry’s internal records were not available at the time. Thus, this work, by examining the industry’s own records on this topic, provides the first comprehensive analysis of the TTC’s own rationale for and tactics used to penetrate new markets. Finally, until now, the tobacco industry documents have largely been studied in isolation from other materials. This thesis provides a unique attempt to contextualise and triangulate the industry documents with other data sources, thereby overcoming potential weaknesses of document research and enabling the document findings to be placed in a wider context.

5.2 Aims

Drawing from the literature review and the gaps in research identified, this thesis aims to understand the impact that the FSU’s transition from command to market economy had on the region’s tobacco industry, on cigarette consumption, smoking prevalence and tobacco control and the role the TTCs played in influencing or benefiting from these changes. In so doing, it aims to inform the development of effective tobacco control policies in this region and in others facing similar changes.

5.3 Objectives

The specific objectives of the thesis are:

113 1. To identify the changes that occurred in the FSU’s tobacco industry with the investment and trade liberalisation that accompanied transition; 2. To examine the impact of these changes on production capacity and tobacco consumption; 3. To describe the prevalence and determinants of smoking in countries of the FSU and seek potential explanations for these patterns; 4. To obtain and analyse internal tobacco industry documents released as a result of litigation in order to: a. examine the interests of the tobacco transnationals in this region and the rationale behind their entry; b. To identify the tactics the TTCs used to gain entry to the countries of the FSU, their responses to and influences on the process of tobacco industry privatisation; c. To understand how the TTCs have sought to influence tobacco control policy makers and policy making throughout the FSU; 5. To combine the findings to assess the impact that transition, including tobacco industry privatisation and TTC entry, had on tobacco control, tobacco control policy making, tobacco consumption and smoking prevalence; 6. To draw lessons from these findings for tobacco control policy.

5.4 The research questions

The research questions posed are thus: 1. What was the scale and nature of the TTC contribution to foreign direct investment in the FSU? 2. How have TTC investments changed the pattern of cigarette production and trade in the FSU? 3. Why and how did the TTCs respond to the opportunities posed by the opening of the FSU and how did they influence the privatisation process? 4. What impact has trade liberalisation and TTC entry had on tobacco consumption?

114 5. What are the current patterns and determinants of smoking behaviour in the region and how might they be explained? 6. How have the TTCs influenced tobacco control policies in the former Soviet Union and how can these influences be explained? 7. Do investment liberalisation and tobacco industry privatisation pose dangers to tobacco control and health? 8. If so, how might these dangers be mitigated?

5.5 Datasets and Methods

Three broad datasets and methodologies will be used in this thesis. The choice is largely pragmatic: they will enable the research questions to be answered. Importantly, each is of value in its own right, whilst also complementing the others, as outlined further below. In brief, these are:

• Identification and analysis of a wide range of data, from published and unpublished sources, on the nature, scale and trends of tobacco industry investments, tobacco trade and consumption. Numerous data sources will be used to obtain as comprehensive picture as possible, with efforts made to critically assess and compare data from different sources in order to provide as detailed and accurate a picture of investment, trade and consumption as possible. • Collection and analysis of survey data in eight of the fifteen countries of the region in order to assess the prevalence and determinants of smoking behaviour. Comparison of the results with data from previous surveys and indirect evidence of historical smoking patterns, inferred from trends in lung cancer mortality will be used to comment on trends in smoking behaviour. • Collation and analysis of internal tobacco industry documents released through litigation in the United States and held in a depository in Guildford, UK. The methods will be described in detail later in the thesis. Rather than presenting a single methodology chapter, this thesis takes a somewhat unusual approach, outlining the methods in four separate chapters: the analysis of routine data is presented in Chapters 2 and 3, the survey methodology in Chapter 4 and the document methodology in Chapter 5. It was felt that it would simplify the presentation of the methods, make the

115 reading of the thesis more straightforward, and allow a constructive discussion of the methods if the relevant methodologies were presented alongside their results.

5.5.1 Combining methods The strength of this thesis lies in its ability to combine these three datasets and methodologies. Just as has been argued in health services research, the problems facing tobacco control in the FSU are multidimensional and require research that draws on a wide spectrum of methodologies. 205 The combination of methods proposed has a number of strengths (Figure 1-11). First, one method helps inform the other. Thus the analysis of TTC investments was necessary to inform the subsequent document analysis: not only did it elucidate which countries each TTC had an important presence in, but it also provided background information on the TTC’s contribution to the country’s economy against which subsequent documentary findings could be interpreted. Second, it allows the weaknesses in one method to be compensated for by the strengths of another. Thus, for example, survey data may show certain patterns in smoking behaviour but they do not help explain why such patterns occur; tobacco industry documents may help provide an explanation. Similarly, the survey data help elucidate the impacts of industry actions that are outlined in the documents. Third, it facilitates triangulation of the data – by having different data sets that examine the same question, albeit from slightly different angles, each can be used both to check the validity of the other and to further elaborate the study findings. Validation is, as will be outlined in Chapter 5, particularly pertinent to the study of tobacco industry documents, where there is a danger that document findings could be taken out of context and/or misinterpreted. The elaboration of study findings using different methods allows a more complete picture of the situation in the FSU to be developed than would be possible using a single methodology alone. The rationale for combining methods in this study therefore has much in common with that commonly advanced for combining qualitative and quantitative methods205 and for combining different qualitative methods, that is, to benefit from each method’s strengths and limitations.206 Barbour makes an interesting point that whilst concerns are often raised with combining of qualitative and quantitative methods on the basis that their epistemological assumptions are incompatible, such concerns are rarely raised when combining different

116 qualitative methods.207 Yet she suggests that within the qualitative paradigm a variety of different methods exist each with distinctive assumptions, frameworks and techniques. It is possible therefore, that many of the problems that have arisen in combining qualitative and quantitative methods have occurred because of the different cultural backgrounds of the researchers; 208 a problem that does not arise when the same researcher is conducting both methods, as in this instance.

117 Figure 1-11 Rationale for combining methods

QUALITATIVE METHODS QUANTITATIVE METHODS

Provides background information on TTVC activity and informs document findings by elucidating potential economic rationale for TTC influence Analysis of investment data

Document findings may help explain patterns and volumes of investment identified

Document findings may help explain trends in trade & consumption Analysis of data on trade Document analysis and consumption Trade & consumption data may help assess impacts of policies/strategies identified in document analysis

Document findings may help explain survey findings Surveys of smoking prevalence and Surveys may help assess impacts of determinants policies/strategies identified in document analysis

Triangulation/validation

118 CHAPTER 2 An overview of industry investments, impact and influence

BACKGROUND When the Soviet Union collapsed, each of the newly independent states inherited its own government-owned tobacco industry. But the centrally funded subsidies for growers and producers had ended and the centralised tobacco import and distribution system broken down, leaving individual factories to fend for themselves. The chaotic state of the industry and the marked cigarette shortages seen at the time74 combined with the rejection of Marxist-Leninist ideology and the lurch towards market reform provided an obvious opportunity for the TTCs; one which they were quick to exploit.

Experience elsewhere shows that entry of the TTCs leads to increased competition with or replacement of the local, unsophisticated industry with a powerful corporation with widespread implications for tobacco control (see Chapter 1, Section 2.4).16,92 Western- style marketing techniques, new lighter brands that are easier to smoke including specific brands for women are introduced. It has also been suggested that state-run industries are less likely than TTCs to deny evidence of tobacco’s health impacts, challenge public health initiatives and obstruct national tobacco control laws by exerting political and commercial pressures.16 In countries with little experience in dealing with such powerful actors, particularly where large sums of money are involved, the impact may be substantial. This is supported by econometric analyses which show that opening tobacco markets to the TTCs through trade liberalisation leads to significant increases in cigarette consumption. 5,98,108

Despite its scale, little has yet been written about the impact of transition and market liberalisation on the tobacco industry in the FSU, which along with central and eastern Europe has been the main focus for tobacco industry investments over the last decade. This chapter examines TTC investments in the region, the contribution these investments have made to total foreign direct investment (FDI) as an indicator of the industry’s economic leverage and the impact TTCs have had on production capacity. It also gives a brief overview of the consequences of TTC investment examining brand development, advertising and tobacco control.

119 METHODS Details of tobacco industry investments in countries of the FSU from 1990 to the end of 2000 were obtained from tobacco industry websites, company annual reports, tobacco industry journals (Tobacco Journal International and Tobacco Reporter), European Bank of Reconstruction and Development (EBRD) investment reports and a review of published literature. The same data sources were used to estimate changes in cigarette production capacity by comparing reported capacity before and after investment. For the latter, the most recent data are reported. Where no post-investment capacity data were available, data on actual production were sought and, if missing, it was assumed that the factory continued to produce at pre-investment levels.

Data on total net FDI inflows to the countries receiving significant tobacco investments were obtained directly from the EBRD for the period 1992 (when the first TTC investments occurred) to 2000. Identical data have since been made available on the United Nations Conference on Trade and Development website (http://www.unctad.org/en/subsites/dite/fdistats_files/fdistats.htm). The proportion of FDI contributed by tobacco industry investments was calculated for each country. Definitions of FDI used were taken from the 2001 World Investment Report where net FDI is defined as credits less debits.209

Information on tobacco control was taken from a number of published sources including the Tobacco Control Database of the World Health Organisation Regional Office for Europe.210

RESULTS

Tobacco industry investments Russia and Ukraine, the two most populous states, received the greatest investment with all the major TTCs staking a claim (Table 2-1,Table 2-2). Reports suggest that foreign- owned factories now account for between 60%211 and 90% of the tobacco market in these countries.212,213,214

The three small Baltic States have each received investments from single companies (Table 2-3). The Danish company, House of Prince (in which BAT has a significant

120 stake) invested in Latvia215 and Swedish Match (then Svenska Tobaks) in Estonia216, each purchasing the country’s only tobacco factory. Philip Morris bought one of Lithuania’s two main factories, and then constructed a second plant, forcing the government-owned plant in Kaunas to close. This has left Philip Morris, Lithuania’s third largest single investor,217 as the sole manufacturer218 with a market share of some 85%.219 TTCs have also invested in three of the five central Asian republics (Table 2-4) securing monopoly positions in two, Kyrgyzstan and Uzbekistan. By 1999, BAT had achieved a market share of over 70% in Uzbekistan211, 220 and as the country’s largest foreign investor 221 , 222 has been able to secure favourable treatment from the government, including a 5-year extension of its preferential tax-exempt status (discussed further in Chapters 9-11).223 While Philip Morris does not have a manufacturing monopoly in Kazakhstan, it nevertheless has a market share of over 70%211 and is the country’s largest taxpayer.224 Conflict and political instability initially delayed investment in the Caucasus, a major leaf producing region in Soviet times. However, between 1997 and 2000, two joint ventures were established in Armenia225,226,227 (Table 2-5), the first with Grand Tobacco of Canada, now reputed to be the country’s second largest taxpayer 228 controlling between 50%225 and 70%229 of the market. Azerbaijan saw failed investments by RJ Reynold’s (now part of Japan Tobacco International) and others, 230 , 231 but in 1999 Athens-based European Tobacco Inc, with support from Altadis, acquired the Baku- factory211,232 for an estimated $50 million.233 By the end of 2000 the five remaining countries in the region had yet to see any direct foreign investment in tobacco. The Georgian tobacco industry was privatised in 1998 and although the TTCs did not invest directly, in 2001 BAT spent a reported US$15 million establishing a licensed production operation in Georgia (personal correspondence G. Bakhhturidze, 2002). In Belarus the arch-conservative President Lukaschenko has resisted privatising the economy, thwarting BAT’s and RJ Reynold’s concerted attempts to establish a joint venture. 234 Turkmenistan has no cigarette manufacturing facility and the highly autocratic President Niyazov, since being advised to give up smoking after major heart surgery, has decreed smoking a vice and banned it

121 in public.235 While Moldova, at least until the election of the Communist Party, had taken a more liberal attitude to market reform, privatisation of its tobacco industry has proved contentious. Despite IMF pressure and purchasing attempts by Reemtsma and BAT the tobacco industry remains in state hands (discussed in more detail in Chapter 8). 236 ,10, 237 , 238 In Tajikistan, the five-year civil war following independence, three changes in government and widespread violence have discouraged investment.239

Contribution to FDI By the end of 2000 the TTCs had invested at least US$2.7 billion in ten countries of the FSU (Table 2-6), US$1.7 billion of this in Russia. The TTC’s contribution to FDI has been most significant in Uzbekistan where BAT’s investment accounts for one third of total FDI. Although the price Reemtsma paid in Kyrgyzstan is unknown, EBRD identify it as one of the country’s major investors and its contribution to FDI is therefore likely to be significant.240 In Russia, Ukraine and Kazakhstan tobacco money accounts for over 4% of FDI, while in the Baltics and Caucasus it has played a less important role. Data for Latvia and Lithuania show that despite Philip Morris being the third largest investor in Lithuania, due to the considerable investments made in other sectors including telecoms and electronics, tobacco investments account for only 1-3% of FDI. Although data are not available for Estonia, the EBRD has not recognised tobacco as a major sector for investment.241

Tobacco FDI per capita (Table 2-6) shows a broadly similar pattern with lower rates in the Caucasus and Baltic states and higher rates elsewhere although Lithuania and Ukraine are exceptions to this pattern.

122 Table 2-1 Details of tobacco industry investments in Russia Date Acquisition Initial Post Initial share, Investment ($) capacity investment % (final (cigarettes) capacity share, %) (cigarettes)* RJ Reynolds (RJR) / Japan Tobacco International (JTI) 1992 RJ Reynolds buy Petro tobacco Factory in St Petersburg and 15bn 50bn 52 (70) Total investment by form the RJR-Petro Tobacco Factory, Russia’s largest 1998 $330 M (Petro production facility factory reported as 1994 RJR buy Yelets tobacco processing plant in central Russia 3000 tonnes NA 49 being acquired for Reconstituted only $25M) tobacco 1995 RJR controlling interest in Armavirtabak, in Krasnodar region 4bn NA NA 1998 RJR invests in RJR-Petro Tobacco Co. to increase production See above See above -- >$120M Philip Morris International (PM) 1993 PM buy plant in Krasnodor southern Russia 2bn 31bn 49 (80) $60M 1993 PM acquire Samara plant in Moscow 4bn NA NA NA 1993 PM build Vyborg plant in St Petersburg. -- 10bn $100 1998 PM reconstruct Krasnodor factory to increase production See above See above $40 Reconstruction completed in 2000

2000 PM open new cigarette factory (Izhora plant) on outskirts of -- 25bn $330M St Petersburg –the largest of PMs 3 Russian facilities. BAT & Rothmans International (which was acquired by BAT in 1999) 1993 Rothmans acquire Nevo plant in St Petersburg 8.8bn 10bn 75 $85 M 1994 BAT buy Saratov plant in Saratov 5bn 10bn (in 75 $40M 1996)

1994 BAT buy Java plant in Moscow second biggest cigarette 11.7bn 30bn 85 $70M factory in Russia

1997 Rothmans announce opening of new cigarette manufacturing -- 10bn $80M plant in Konnaya Lakhta industrial zone north-west of St Petersburg 1997 BAT modernises its cigarette plants in Java & Saratov Russia See above See above See above >$150 1998 Further investment approved for Java and Saratov See above See above See above $65M Gallaher / Liggett-Ducat (which was bought by Gallaher in 2000)

123 1993 Liggett Brooke acquire Ducat plant in Moscow and establish 10.5bn 20.3bn (in 70 NA Liggett- Ducat 1998) ($4m later invested) 1999 Liggett-Ducat inaugurates new cigarette manufacturing plant -- >40bn $85M in Moscow 2000 Gallaher reported to have further invested in this plant -- -- $60M Others 1998 Reemtsma purchase majority share in cigarette factory in NA Planned NA $100M Volgograd 20bn by 2002 Totals 61 260.3 >$1719M 212,66, 267,242,243, 83, 244, 6, 245, 246, 247, 248, 249, 250, 251, 252, 253. Sources: * where figures for post investment production are not available, it is assumed that pre-investment levels apply NA= Not available a Market share data: Corrao et al

124 Table 2-2 Details of tobacco industry investments in Ukraine Date Acquisition Initial capacity Post investment Initial share, % Investment ($) (cigarettes) capacity (cigarettes)* (final share, %)*** RJ Reynolds (RJR) / Japan Tobacco International (JTI) 1992 RJR buy Livovskaya plant in Lviv 11bn 0 (closed) 70 $9.6m initial investment 1992 RJR buy Tabaknaya plant in Kremenchuk 8bn 12.5bn 70 (92) $7.3M initial investment BAT 1993 BAT establish Joint venture, the BAT Prilucky 12bn 16bn 65 (75) $65M Tobacco Company Reemtsma 1993 Reemtsma buy factory in Cherkassy # 10bn 12.6bn 65 (92) total $55M (120M 1995 Reemtsma buy factory in Kiev 5bn See below 65 (87) DM) since 1993

1998 Reemtsma open new factory in Kiev and -- 15.5 relocate production from old site Philip Morris International (PM) 1994 PM buy Kharkov Tobacco Factory 2.5bn 9.4bn 51 (99.9) Undisclosed amount. Additional investment of $16M later announced Totals 48.5bn** 65.4bn >$152.9M

214,242,254,255,238,256,257,258,259 Sources: * Where figures for post investment production are not available, it is assumed that pre-investment levels apply ** Although capacity is estimated at 48.5 billion, actual production was well below this at an estimated 34.3billion *** Additional data on final market share provided by Konstantin Krasovsky # In 2001, the Gallaher group bought the Cherkassy factory from Reemtsma and it will now be operated by Gallaher’s subsidiary Liggett-Ducat260 NA= Not available

125 Table 2-3 Details of tobacco industry investments in the Baltic States Date Acquisition Capacity Post investment Initial share, % Investment ($) (cigarettes) capacity (cigarettes) (final share) ESTONIA 1993 Swedish Match (then Svenska Tobaks) 5.3bn (production 2bn) 5.3bn (production 3bn) in 51 (65) NA invest in Leek tobacco factory and create 1996. Factory then Eesti Tubakas# closed. LATVIA 1992 House of Prince established a joint venture 5bn (production 2.6bn) 5bn (production 6.4 bn) 45 $12.5M House of Prince Riga (and investment of $ 15M planned) LITHUANIA 1993 PM buy controlling interest factory in 3bn 3bn 65.2 $62M Klaipeda Tobacco Company 1995 PM begin work on new manufacturing plant -- 6bn in Klaipeda (production begins 1997) 215, 217,261,262 Sources: # In 1999 Swedish Match sold its tobacco business to Austria Tabak which was in 2001 taken over by the Gallaher’s Group. NA= Not available

126 Table 2-4 Details of tobacco industry investments in Central Asia Date Acquisition Initial capacity Post investment Initial share, Investment ($) (cigarettes) capacity (cigarettes)* % (final share, %) KAZAKHSTAN 1993 PM buy controlling interest in Almaty Tobacco Company 12bn See below 49 (99) $340M total PM 2000 PM modernise existing factory & open new cigarette -- 25bn investment manufacturing plant in Almaty. 1993 RJR buy former chocolate factory in Shimkent and turn it -- NA 97% $100M into a modern tobacco facility producing Winston and Camel 1997 Reemstma and Gallaher jointly establish a tobacco factory NA NA NA NA 25km from Almaty Totals 12bn >25bn >$440M KYRGYZSTAN 1998 Reemtsma acquire Kyrgyzstan’s only cigarette production 0.85m NA NA NA plant and establish JV with government 2000 Reemtsma open new production site in Bishkek -- 8bn NA NA Totals 0.85 bn 8.85bn NA UZBEKISTAN 1994 BAT buy UZ plant in Tashkent and create UzBAT 5bn (output 2bn) 12bn by mid-1996. 51 (97) >$300M Production then transferred to new factory 1996 Construct factory in Samarkand -- 30bn Totals 5bn 30bn $>300M

222,242,263,224,264,265,266 Sources: * Where figures for post investment production are not available, it is assumed that pre-investment levels apply NA= Not available

127 Table 2-5 Details of tobacco industry investments in the Caucasus Date Acquisition Capacity Post investment Initial share, % Investment ($) (cigarettes) capacity (cigarettes) (final share) ARMENIA 1997 Grand Tobacco of Canada forms joint venture 5bn 6bn NA $8M with two local companies (RRR and Samsum) 2000 Greek company forms joint venture with -- -- NA NA tobacco processing plant (Masis-Tabak) AZERBAIJAN 1999 European Tobacco forms JV, European NA NA NA $50M Tobacco-Baku 225,226 Sources: , NA= Not available

128 Table 2-6 Total net foreign direct investment (FDI) and tobacco investments as a proportion of total FDI. Figures in US$ millions

Russia Ukraine Kazahkstan Kyrgyzstan Uzbekistan Armenia Azerbaijan Estonia Latvia Lithuania TOTAL all 10 countries

GNP per capita (US$) 1,690.00 690 1,260.00 280 620 520 600 3,580.00 2,920.00 2,930.00 Population size (1995) 147,773,664 51,276,560 16,539,600 4,481,611 22,912,100 3,759,950 7,444,318 1,483,942 2,515,602 3,714,795 261,902,142 Total FDI 1992 700 200 100 40 2 82 27 10 1161 Total FDI 1993 1211 200 1271 10 45 1 0 162 45 30 2976 Total FDI 1994 640 159 660 38 50 8 22 215 214 31 2036 Total FDI 1995 2016 267 964 96 120 25 282 202 180 73 4224 Total FDI 1996 2479 521 1137 47 55 18 661 151 382 152 5602 Total FDI 1997 6638 624 1321 83 285 52 1093 267 521 355 11238 Total FDI 1998 2761 743 1152 109 140 232 1024 581 357 926 8024 Total FDI 1999 3309 496 1587 35 121 130 510 305 348 486 7328 Total FDI 2000 2704 595 1249 19 100 133 500 398 407 379 6484 Total FDI 1992-2000 22,458 3,804 9,441 438 956 601 4,092 2,361 2,480 2,442 49,073 Country making greatest USA (34.0%) USA (16.45%) USA (28.4%) USA (18.6%) n/a n/a n/a Sweden USA Denmark contribution to total FDI (% of (1993-1999) (1997-1999) (39.2%) (11.8%) (16.1%) total) over the period 1993 - 2000 (unless otherwise stated)

Tobacco recognised by EBRD No Yes: Rma, Philip No Yes: Rma Yes: BAT Yes: Grand No No No Yes: PM report as a major sector for Morris, BAT Tobacco & listed as 3rd investment? Masis-Tabak largest investor

Total reported FDI in tobacco >1719 >152.9 >440 n/a >300 >8 50 n/a 12.5 62 >2744.4 sector to end 2000

Tobacco investments as a % >8% >4% >5% n/a >31% >1% 1% n/a 1% 3% >6% of total FDI Tobacco FDI per capita >11.63 >2.98 >26.60 n/a >13.09 >2.13 6.72 n/a 4.97 16.69 >10.48 PM=Philip Morris, BAT = British American Tobacco, Rma = Reemtsma Data sources: WHO Health for All database (population data, GNP), European Bank of Reconstruction and Development Investment Profile reports, 2001 (FDI data) and other data as given in tables above. * If Kyrgyzstan and Armenia, on which tobacco FDI data are unavailable are excluded, the results are almost identical - tobacco investments still contribute >6% of total FDI and tobacco FDI per capita is >10.72.

129 Table 2-7 Percent market share for 1998-2000 (or nearest years where data available) based on legitimate sales only for all countries bar Belarus Azerbaijan Belarus Estonia Georgia Kazakhstan Kyrgyzstan Latvia Lithuania Moldova Russia Ukraine Uzbekistan 1998 1999 2000 1999 2000 1999 2000 1998 1999 2000 1998 1999 2000 1998 1999 2000 1999 2000 2000 2000 1998 1999 2000 2001 1998 1999 2000 1999 2000 Multinational manufacturers BAT 6.4 10.1 11.9 13.1 13.2 13.3 12.6 10.1 16.8 71.9 50.9 Philip Morris 51.4 63.0 70.0 78.0 75.0 45.0 47.0 84.0 28.5 18.4 18.5 19.5 20.0 18.2 22.6 JTI/RJR 15.0 12.0 8.5 5-10 18.0 19.2 20.0 20.0 15.0 5.0 12.0 Gallagher/Liggett Ducat/Austria Tabak 28.6 27.0 10.0 4.2 6.9 minor 7.3 10.0 9.6 12.3 Reemtsma 17.0 35.0 51.0 5-10 2.0 38.0 38.5 34.0 House of Prince Riga (BAT associate) 35.0 34.0 5-10 Seita/Altadis minor sub-total 0 0 0 0 0 86 100 0 0 0 95 94.2 90.4 17 35 51 80 81 84 0 66 61 61 67 86 72 85 72 51 Other named manufacturers Amer Tobacco 5.1 Bulgartabak 2.5 2.3 2.0 Balkan Star 7.4 CNTIEC-Kazakhstan Donskoy Tabak 9.0 European Tobacco Baku 33.9 64.4 64.0 Neman 40.0 39.0 Nevo Tabak 2.6 Tabak Invest 2.0 6.0 Tutun CTC 93.6 Others Other manufacturers 1.2 2.9 8.5 20.3 36.9 44.3 3.0 1.6 Other importers (includes TTCs) 64.9 32.7 36 15.0 15.0 79.7 63.1 55.7 83 65.0 49.0 4.8 Others unspecified 5.0 5.8 6.6 20.0 19.0 34.3 36.8 36.4 11.9 14.4 28.2 14.6 28.1 49.1

Smuggled / counterfeit brands (where data given - Belarus only) 43.0 40.0

Source: ERC statistics Note: no data given at all for Armenia or Tajikistan and no market share data for Turkmenistan in part due to the high levels of smuggling.

130 Table 2-8 Tobacco control policies by country Armenia Azerbaijan Belarus Estonia Georgia Kazakhstan Kyrgyzstan Latvia Lithuania Republic of Russian Tajikistan Turkmenistan Ukraine Uzbekistan Moldova Federation Direct advertising National TV Complete Complete partial Complete partial partial partial Complete Complete Complete Complete none Com plete Complete partial Cable TV Complete Complete partial none partial none partial Complete Complete none Com plete none Com plete Complete none National Radio Complete Complete partial Complete partial partial partial Complete Complete Complete partial none Com plete Complete partial Local magazines, newspapers Partial Complete partial Complete partial none partial partial Complete partial partial none Com plete partial partial International magazines, newspapers None Complete none none none none partial none none none none none none none none Billboards, outdoor walls none Complete partial Complete partial none n/a Com plete Complete none none none none partial partial Points of sale, kiosks none Complete none Com plete none none none none Com plete none none none none none none Cinema none Complete none Com plete none none n/a none Com plete none partial none none none none Indirect advertising Product placement TV and films none Complete none Com plete partial none partial Complete Complete none partial none none none none Sponsored event with tobacco brand none Complete none none none none none none Complete none none none Voluntary partial none name agreement Non-tobacco products with tobacco none n/a none n/a none none none none Complete none none none none partial partial brand names Non-tobacco product brand name used none n/a none Complete none none n/a none Complete none none none none none n/a for tobacco Direct mail giveaways none Complete none n/a none n/a n/a none Com plete none none none none none none Promotional discounts none partial none n/a none n/a n/a none Com plete none none none none none none Distribution of tobacco products through various outlets Vending machines none Complete none Com plete none none n/a Com plete Complete Complete Complete none none Com plete none Self-service displays none partial partial none none none none none none Com plete none none none partial none Mail order or electronic sales none none none none none none none none none Complete none none none none none Sale of single or unpacked cigarettes none Complete none Complete none none none partial Complete Complete Complete none none Complete none Sales of duty free tobacco products none none partial none none none n/a none none Com plete none none none none none Free samples of cigarettes none Complete partial none none none none Com plete Complete Complete none none none Com plete none Smoke free areas Health care facilities partial Complete Complete Complete partial partial Complete Complete Complete Complete Complete none Com plete Complete partial Education facilities none Complete partial Complete partial none Com plete Complete Complete Complete Complete none Com plete Complete partial Government facilities none Complete partial Complete none partial none partial partial partial Complete none Com plete partial none Restaurants none partial none partial none none none partial partial partial none none none none none Pubs and bars none none none partial none none none partial none none none none none none none Indoor workplaces and offices none Complete partial Complete none none partial partial Complete partial Complete none Com plete partial none Theatres and cinemas none Complete partial Complete partial none none partial Complete partial Complete none Com plete partial partial Smoke free public transport Buses none Complete Complete Complete partial none Com plete Complete Complete Complete Complete none Com plete Complete partial Taxis none none Complete Complete none none none Com plete Complete Complete none none Com plete partial none Trains none Complete partial Complete none none n/a partial partial partial partial none Com plete partial none International air transport none Complete partial Complete partial none n/a partial Complete Complete partial none Com plete partial Complete Health warnings Health warning on tobacco products Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes Yes

Source: WHO Regional Office for Europe, Tobacco Control Database http://cisid.who.dk/tobacco/ (last accessed 12 May 2003)

131 Consequences of TTC Investments Production capacity and market share Production capacity has increased ten-fold in Kyrgyzstan, six-fold in Uzbekistan, over four-fold in Russia, tripled in Lithuania and doubled in Kazakhstan (Table 2-1-Table 2-5). In Ukraine, Estonia and Latvia increases in capacity have been more modest although actual production has increased by at least 50%.

According to data on legally traded cigarettes, TTC market share has increased from zero pre-1990 to reach levels of 50-100% in markets in which they have invested (Table 2-7). However, levels of smuggling remain substantial, particularly in countries where the TTCs have yet to invest directly,211 and the TTC’s market share is therefore likely to be larger in these markets than official data suggest.

The introduction of branding and advertising In the Soviet era, the western concept of branding was virtually unknown. Almost all cigarette brands were state-owned and each factory produced a variety of brands (the exception was the Yava brand, produced exclusively by the Yava factory later acquired by BAT). Most were traditional filterless cigarettes (papirossi or oval cigarettes) with a few low-priced filter brands.66 Since the mid 1990’s the TTCs have gained ownership of existing brands, developed new brands specifically for these markets and introduced their own international brands. JTI, for example, introduced eight new brands in 1999 alone.267 The TTCs have also focused production on filter brands, the consumption of which increased through the 1990s, although this trend reversed temporarily after the 1998 financial crisis.214,268

The Soviet Union had a wide range of anti-smoking policies. Although advertising was unnecessary and so non-existent, tobacco advertising was banned, smoking was forbidden in many public places including subways, buses and restaurants, cigarette packages carried health warnings and anti-smoking campaigns were televised.70,68 Post- transition the TTCs exploited confusion over the legality of this Soviet legislation by advertising heavily to establish their brands84 leading to a major surge in advertising and promotion.86 By the mid 1990s it was estimated that up to 50% of all billboards in Moscow and 75% of plastic bags in Russia carried tobacco advertising.63 The TTCs were soon identified as the largest advertisers on Russian television and radio and in at

132 least four of the former Soviet states the tobacco transnationals ranked amongst the top 3 advertisers.269,270 With the advent of television advertising bans, for example in Russia, industry spend shifted to other media - tobacco is now the product most heavily advertised outdoors with three major transnationals ranked as first, second and third heaviest advertisers.271 An almost identical pattern is seen in Ukraine and Belarus.271,272

Tobacco control Tobacco control policies have been enacted to varying degrees in each country (Table 2-8). Initial attempts to implement effective tobacco control policies were fraught with difficulties. In Russia, the absence of enforcement mechanisms rendered the 1993 tobacco advertising ban ineffective.128 In 1995 a further law was passed, but based on the industry’s voluntary code of conduct273 only included minor restrictions on content, placement of outdoor adverts and timing of broadcast adverts (see also Chapter 11).274 In July 2001 a new federal bill on Limitation of Tobacco Consumption was signed, to be introduced in stages from 2002. The original bill included highly effective measures but, as the St Petersburg Times notes, the changes made to the draft between the first and second readings were “a textbook demonstration of the lobbyist’s art”.275 The initial tobacco advertising ban was excluded when the industry successfully argued that it should be subject to a separate law. Similarly the original ban on showing smoking in movies has been weakened to allow smoking if it is “an integral element of the artistic design.”275 Some restrictions on smoking in public are set but no clear system of enforcement is specified. While a few of the remaining components of the bill remain useful, in particular a ban on sales from vending machines, of packs of under 20 and single cigarettes, it is clear that industry interests triumphed.276 Similar problems have been encountered elsewhere. In Ukraine a 1994 decree failed to define “advertising” and was therefore easily contravened. In 1996 the Rada gave preliminary approval to a law that was essentially a translation of the voluntary code submitted by the industry but, following protests from the tobacco control lobby, ultimately enacted a stronger ban.14 The industry lobby responded with a report from the so-called “Association of Independent Advisors”, for which Philip Morris later admitted responsibility, arguing that Ukraine would lose US$400 million as a result of an advertising ban.14 This led to a Presidential veto of the ban and finally, in July, the Rada

133 introduced a much weakened ban that covered only television, radio and cinema advertising. As in Russia, there are many loopholes and enforcement is weak. Ukraine, however, has a more active tobacco control lobby than many other FSU countries and a recently established coalition of anti-smoking organisations has made some progress in promoting more effective tobacco control policies. The central Asian states Kazakhstan, Kyrgyzstan, Tajikistan and Uzbekistan have few measures to reduce tobacco use; in contrast in Turkmenistan progress has been made because of the strong position taken by the president.277 The picture in the Caucasus varies widely with excellent tobacco control policies officially in place in Azerbaijan, but weak policies in Armenia and Georgia where implementation of an advertising ban approved in 1998 was delayed by industry lobbying. Progress in Moldova and the three Baltic States has been better. Lithuania has a strong public health lobby in part as a result of individuals participating in international research projects even during the Soviet era.131 It was one of the first east European countries to develop comprehensive tobacco control legislation 278 and has been described by industry journals as “the most virulently anti-smoking in the Baltic States”. 279 Nevertheless, its advertising ban, which covered both direct and indirect advertising, and was due to come into force in July 1996, was not implemented until May 2000 due to a series of challenges. Estonia enacted its first advertising ban in 1993 but the industry exploited loopholes in the law, requiring a further act which covers direct advertising (other than cable TV and international print media) but which does little to restrict indirect advertising. The second act came into effect in 1998 and a subsequent Tobacco Act, banning smoking in public places and smokeless tobacco, was adopted in June 2000. Latvian attempts to pass an advertising ban in 1996 failed due to intense opposition. A subsequent ban has outlawed television, radio and outdoor advertising but does not restrict point of sale, cinema or indirect advertising and only partially restricts adverts in the printed media. Industry journals do not hide the fact that the tobacco companies were angered by the tobacco control measures taken in these three countries, in particular the substantial excise tax increases (the three Baltic States have considerably higher excise rates than most other FSU countries)280 and the governments’ refusal to maintain import duties to

134 protect investing companies.219,281 The latter seems somewhat ironic given the industry’s frequent lobbying for reductions in trade barriers where such changes suit their interests. The industry claims that the Baltic governments have contravened promises they made to introduce favourable taxation rates and, as frequently argued elsewhere, that price increases have led to an influx of cheap smuggled cigarettes that make it impossible for them to compete.281,282

DISCUSSION While the data provided in this chapter are as comprehensive as possible, a number of potential weaknesses must be considered. Firstly, data on FDI are notoriously difficult to collect. For example, assigning values to investments that include both equity and other forms of capital investment (machinery etc) can be complex and will affect both the total and tobacco investment figures used in this chapter. In some instances I was unable to obtain data on tobacco industry investments and in other instances only minimum investment figures were available. Tobacco industry investment will therefore tend to be underestimated. However, due to the high levels of total FDI compared to levels of tobacco industry investment, even relatively large increases in tobacco industry investments would have only a small impact on the calculated percentage contribution to FDI. Thus for example, anecdotal reports suggest that Philip Morris invested $60 million (I did not include this figure as I was unable to verify it) and the TTCs together a total of approximately $200 million in Ukraine (compared to my estimated total of $153 million which excluded the Philip Morris investment).283,284 This would however only increase the tobacco industry’s contribution to FDI from our estimate of over 4% to 5.3%. Despite these weaknesses, the data provide the first comprehensive assessment of industry investments in the FSU and are useful in providing some assessment of the relative contribution of tobacco to total investments country by country. The data presented suggest that production capacity in the countries receiving investments has almost tripled. Actual output may have increased by a greater amount as many factories operated well below capacity in the pre-transition period and post- investment capacity is unknown in some instances. For example, I was unable to identify new capacity levels in Latvia but United States Department of Agriculture data suggest

135 production increased from approximately 2.6 to 6.4 billion cigarettes between 1993285 and 2000.286

Transition clearly heralded changes to the region’s tobacco industry that will have major implications for future health. The TTCs established new cigarette brands, exerted ownership over existing brands, expanded marketing and used their powerful lobby to undermine the development of effective tobacco control. The argument that foreign investment simply led to a substitution of traditional tobacco products with new ones can therefore be comprehensively rejected. Moreover, despite difficult trading conditions and the 1998 economic crash, the TTCs already report good returns on their investments.287,288,289,290,291,292 The findings suggest that, in the fifteen countries of the FSU, five distinct patterns can be ascertained, corresponding to the size of the country, the nature of the political and economic transition and the emerging political situation in each. The latter varies along a spectrum from parliamentary democracy at one end, best exemplified by the Baltic States, to personal dictatorships at the other with Turkmenistan the clearest example.293 These factors were important both in influencing initial TTC interest and investments and determining their degree of influence once investment had occurred. First, Russia and Ukraine attracted substantial inward investment in many sectors, reflecting the extensive privatisation of the existing broadly-based industrial infrastructure. The TTCs competed for market share in what were the FSU’s largest potential markets, with none achieving dominance. The strength of the industry, coupled with the lack of non-governmental organisations and low awareness of the health consequences in weak legislatures, meant that effective action was undermined or delayed. However, a strengthening civil society and fledgling tobacco control lobby is now emerging with some progress seen in tobacco control. The second model, seen in the Baltic States, is characterised by some inward investment but only moderate political influence by the industry which, to some extent, has been countered by the tobacco control community. The small population sizes make these markets less important to the industry and tobacco is a much smaller part of the overall economy. Thus the scope and incentive for industry influence is smaller. But perhaps more important is the development of open, pluralistic political systems, with lower

136 levels of corruption and better developed civil society. Their planned accession to the European Union has also been a factor, although this has been used by both industry and tobacco control advocates, with Estonia enacting legislation based on EU policies but tobacco lobbyists in Lithuania citing the annulment of the EU advertising ban as a reason to weaken and delay advertising legislation. The third model is seen in Uzbekistan, Kyrgyzstan and Kazakhstan, countries with autocratic governments that have seen large industry investments and the establishment of monopoly (Uzbekistan and Kyrgyzstan) or near monopoly positions (Kazakhstan) despite Uzbekistan and Kazakhstan being, respectively, the third and fourth most populous post-Soviet states. The TTC’s major contribution to FDI gives them considerable political and economic power, power that appears to be easily wielded in these highly centralised, one party states. Moreover, given that dissent and free speech are poorly tolerated, tobacco control groups are virtually unknown and tobacco control policies are amongst the weakest in the region.294 A fourth group of countries have avoided tobacco investments, at least until recently, and tobacco control has made measured progress most notably in Moldova, the most democratic of the four, and Azerbaijan. The lack of investment reflects resistance to economic reform in general (Belarus and Turkmenistan), more specifically to privatisation of key sectors deemed to be of national importance (Moldova), or to chaotic post-transition conditions that discouraged and undermined attempted investment (Azerbaijan). The Turkmen president occupies a position of almost unparalleled power in his country295 and tobacco control progress there can be attributed his personal animosity towards tobacco. Belarus is still considered an important market by the TTCs, who hope that integration with Russia will occur.296 As a result, BAT and Philip Morris are amongst the largest advertisers in Belarus271 despite their small official market share211 and progress on tobacco control has been limited. Some sectors of the Azerbaijani economy have opened up to enable exploitation of its substantial oil reserves and although Azerbaijan recently received inward tobacco industry investment, it had previously enacted tough anti-smoking policies, at least on paper.

Finally, there is a group of countries affected either by war or almost complete economic collapse, in which there are few functioning institutions and very limited governmental

137 capacity (the Caucasus and Tajikistan). Despite obvious difficulties, smaller international tobacco companies have recently invested in Armenia. The industry seems to regard Tajikistan as too difficult to invest in and perhaps better served by exports from elsewhere. Outside the FSU a similar situation exists in Albania. These countries have weak tobacco control policies, which seems to reflect the lack of governmental capacity rather than direct TTC influence. While there are many similarities with TTC entry into Asia, some underlying features are different. Firstly, liberalisation of inward investment has ultimately been more important than trade liberalisation. Secondly, market entry coincided with huge political and economic upheavals, with legislative activity focused on basic state building, the need to develop a constitution and implement economic reform. There were no effective tobacco control policies in place to act as a buffer against the industry and the development of new legislation was understandably given a low priority amidst other pressing demands. The entry of the TTCs, with their millions of investment dollars, was a further disincentive to effective tobacco control. Moreover, civil society was not well developed and proponents of tobacco control had little voice. As the industry journal World Tobacco for Russia and Eastern Europe stated with delight, “anti-tobacco activists are almost unknown in Russia so the Russian people and government have not been bombarded with anti-tobacco propaganda.”63 As a result, TTCs met with little if any resistance, unlike their experience in Thailand (see Chapter 1, Section 2.4). Indeed investments were often encouraged, not only by governments but also by international financial organisations such as the IMF.237,297 Thirdly, the ability to take over existing monopolies in all but the largest countries contrasts with the competitive positioning seen in Asian markets. Together with their major contribution to FDI, this has given the tobacco companies a unique degree of political influence. In well-functioning democracies, such influence may be effectively counteracted, as illustrated by the experience of the Baltic states. But elsewhere in the FSU, particularly in the central Asian republics, industry and government collusion has left the industry in an extremely powerful position. This will be examined further in Chater 9 to 11 which focus on the situation in Uzbekistan.

138 The following chapter examines the changing pattern of tobacco consumption in this region in the light of the TTC investments.

139 CHAPTER 3 Exploring the impact of foreign direct investment on tobacco consumption in the former Soviet Union

BACKGROUND As described in Chapter 1, Section 2.4, previous studies show that the forced entry of the TTCs into the Asian markets in Taiwan, Korea, Thailand and Japan under the threat of US trade sanctions298 led to an increase in per capita consumption of about 10%.98 Other econometric studies show that greater trade openness (measured using total trade as a share of gross domestic product and import penetration) has a significant and positive impact on tobacco consumption that is greatest in low income countries.108,5 Such findings are consistent with economic theory which suggests that reducing trade barriers increases tobacco consumption through an increase in both supply and demand, the latter driven by and through competition, which reduces prices and increases advertising expenditure.108 The TTCs entry to new markets and the surge in global trade of tobacco products since the 1980s has been enabled by trade liberalisation, driven by bilateral, regional and multilateral agreements that have reduced both tariff and non-tariff barriers.5

However, in addition to exporting to a foreign market, companies can access new markets by establishing or acquiring the facilities to produce in-country and sell directly to the domestic market. Over the last decade such foreign direct investment (FDI) has grown considerably faster than trade, leading some to argue that ‘globalisation of production’ now outweighs ‘globalisation through trade’ in economic importance.4 It offers TTCs the advantage of accessing cheaper labour, inputs and transport costs, avoiding import duties and developed world regulations on disposal of cigarette production waste.299 In addition, privatisation can be expected to improve productivity and efficiency, which if lowered costs are transferred to consumers, could lead to lower- priced products. Yet to my knowledge only one attempt has been made to explore the impact of FDI on tobacco consumption.108 It suggested that an increase in exchange rate distortions (used to indicate a disincentive to investment) led to a decline in cigarette consumption, leading to the tentative conclusion that FDI should lead to higher levels of cigarette consumption. Certainly the theoretical impact of FDI, in terms of its consequences for supply and demand, is likely to be similar to that of trade

140 liberalisation. In addition FDI gives the transnationals additional economic and political leverage within the country concerned.6

As shown in Chapter 2, between 1992 and 2000 the TTCs invested over $2.7 billion in the tobacco industries of ten of the fifteen FSU states accounting for between 1% and over 31% of the total FDI in these countries. This led to major changes, including the introduction of branding and advertising, which were previously unknown. The newly created countries were in the process of developing their own constitutions with new legislative and taxation systems, so none had in place, nor was able to rapidly enact, tobacco control laws. Nor did they have established tobacco control or civil society groups to oppose industry pressure.

Despite the scale of these changes, little is known about their impacts. This chapter therefore seeks to explore the impact that foreign direct investment has had on patterns of cigarette trade and, in turn, on cigarette consumption in the FSU. In so doing it aims to add to the growing body of evidence on the impact that trade liberalisation and transition from a socialist to a market economy has on health. Given evidence that the International Monetary Fund (IMF) is pressuring countries to privatise their tobacco industries and making privatisation a prerequisite for loans,237,10 it is becoming increasingly important to understand what impact privatisation might have. The economic turmoil accompanying transition, periods of rapid inflation, introduction of new currencies and redenomination of old ones, makes interpretation of financial data including cigarette prices across these fifteen countries extremely difficult so this chapter takes a descriptive rather than an econometric approach.

METHODS Three main data sources were used, the United Nations Food and Agriculture Organisation (FAO) database which provides data from 1961 onwards,300 the United Nations Commodity Statistics Yearbooks which provide cigarette production data from 1963 and the United States Department of Agriculture, Foreign Agricultural Service (USDA, FAS) data which are available from 1960.301 The accuracy and completeness of the data were compared with each other and with other sources in order to identify the most appropriate source for each measure of interest (Table 3-1).

141

Table 3-1 Data sources and details Data Data source Year (boundary for import or export data) Tobacco leaf UN FAO database 1961-2000 production (agricultural production section)302 Cigarette USDA 1960-2001 production Tobacco leaf UN FAO database 1961-1995 using USSR imports and (agriculture and food boundaries exports trade section)303. 1992-1999 using NIS boundaries Cigarette UN FAO database 1961-1995 using USSR imports and (agriculture and food boundaries exports trade section) USDA. 1960-1989 using USSR boundaries 1990-2001 using NIS boundaries Cigarette USDA 1960-2001 consumption Mid year United Nations 1960-1989 population Demographic estimates Yearbooks304,305,306 WHO Health for All 1990 –2001 database117 Mid year WHO Health for All 1990-2001 population database117 15+ estimates:

All data are presented for the region as a whole, the USSR until transition and the former Soviet Union (FSU) as a whole post transition. The demise of the FSU does not present problems when examining production or consumption data over time, with data simply aggregated where necessary. It does, however, lead to potential difficulty when comparing import and export data as products traded between different parts of the USSR did not, until the collapse of the USSR, contribute to international trade figures. The FAO database allows for this transfer by providing trade figures for the old boundaries (i.e. for the USSR) up until 1995 and for the new boundaries from 1992 to 1999 giving a four year period of overlap. By contrast USDA simply provides data for

142 the old boundaries up to 1991 and for the new boundaries from 1992 (personal correspondence Arnella Trent, USDA). For trade figures I therefore present both sets of data up to 1995 in order to examine the impact that these changes in configuration had.

Where not already provided, cigarette consumption was calculated from USDA data using the formula: production + imports – exports. Consumption per capita was calculated for the population as a whole using mid-year population estimates from the United Nations Demographic Yearbooks for the years to 1990304,305,306 and the WHO Health for All database (which uses data from the United Nations Population Division) for the years 1990 onwards.117 Whole population data were used rather than the population aged 15 and over as accurate data on the latter were not available across the whole time period. For the period 1991 onwards I examined consumption per capita using the population aged 15 plus (the conventional way of assessing tobacco consumption).

The newly independent states can be split into two groups, those without direct industry investments (Belarus, Georgia, Moldova, Tajikistan and Turkmenistan) and those with substantial investment from the tobacco transnationals in the early to mid 1990s (Latvia, Lithuania, Estonia, Russia, Ukraine, Kazakhstan and Uzbekistan (see Chapter 2)). Trends in tobacco leaf imports, cigarette production and consumption were compared in these two groups of countries. Kyrgyzstan, Armenia and Azerbaijan were excluded from these analyses because although they have now received investments from the tobacco industry this only occurred after 1997, considerably later than the other countries, and it was felt that insufficient time had elapsed for these investments to have had an observable impact.

143 RESULTS Tobacco leaf production Agricultural production of tobacco in this region has varied markedly over time with a drop in the late 1970s and early 1980s, a peak in the mid-1980s, followed by a marked decline until the mid-1990s (Figure 3-1). This recent decline is consistent with reported shortfalls in tobacco during this period. It appears to reflect a number of factors307 including policies to discourage production as part of Gorbachev’s health campaign in the 1980s, droughts and wars, and the demise of Soviet subsidies for agricultural production. 308 Gorbachev’s health campaign focused largely (and effectively) on reducing alcohol consumption,309 but some believe it also aimed to reduce cigarette consumption through reducing supply of leaf and manufactured cigarettes. Others, however, have suggested that the campaign really only served to hide the underlying economic difficulties that were driving down production.

In the mid-1990s production stabilised and now appears to be increasing although it remains lower now than at any time since 1961. The traditional tobacco producing areas of Moldova, Azerbaijan and Kyrgyzstan274 are all recovering from slumps in production post-transition although little increase in production has yet been seen in Azerbaijan.310 Interestingly, production has also increased in countries that have not traditionally been major tobacco producers, notably Uzbekistan and Kazakhstan, reflecting foreign investment by British American Tobacco (BAT) and Philip Morris respectively in their leaf growing industries.311,312,224 (Figure 3-2). However, as we explore further in chapter 7, these increases were not as substantial as the TTCs had led governments to believe.

144

Figure 3-1 Tobacco leaf production in the USSR/FSU, 1961-2000.

450,000

400,000

350,000

300,000

250,000

200,000

150,000 Tobacco leaf production (metric tons)

100,000

50,000

- 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Year Source: UN FAO database agricultural production

145 Figure 3-2 Tobacco leaf production in selected countries, 1992-2000

60000

50000

40000

Azerbaijan, Republic of Kazakhstan Kyrgyzstan 30000 Moldova, Republic of Tajikistan Uzbekistan

20000 Tobacco leaf production(metric tons)

10000

0 1992 1993 1994 1995 1996 1997 1998 1999 2000

Source: UN FAO database agricultural production

146 Cigarette production Cigarette production fluctuated from 1960 with a slow, overall upward trend that peaked in 1986 (Figure 3-3). The rapid subsequent decline has been attributed variously to obsolete manufacturing equipment, shortages of raw materials (tobacco leaf, paper and filters) and, once again, although with little supporting evidence, Gorbachev’s health campaign. Since the mid-1990s cigarette production has increased almost exponentially and has now reached higher levels than ever previously seen with a 76% increase between 1991 and 2000. Production in countries receiving foreign investment increased by 96% during this period, compared with only 11% in countries not receiving investment (Figure 3-3).

Figure 3-3 Cigarette Production in the USSR/FSU, 1960-2001

600,000 FSU

Countries without investment

500,000 Countries with investment

400,000

300,000

200,000 Cigarette production (millions of units)

100,000

0

0 2 0 8 0 8 64 72 82 92 00 974 976 984 196 196 19 1966 1968 1970 19 1 1 1978 198 19 1 1986 198 199 19 1994 1996 199 20

Source: USDA data. Notes: 2001 data are estimates.

147 Imports and exports Imports of cigarettes fluctuated, albeit displaying an overall upward trend between 1960 and 1984 (Figure 3-4). A rapid decline then occurred throughout the rest of the 1980s. In 1990 and 1991 imports suddenly rose due to the airlift into the USSR of a reported 34 billion manufactured cigarettes by Philip Morris and RJ Reynolds.74 USDA data suggest that imports then increased steadily between 1993 and 1995, declining rapidly thereafter. Importantly, this temporary increase was seen only in those countries where the transnationals had invested. These patterns are consistent with the production data described above, suggesting that imports increased until local production picked up from 1995 onwards. Prior to transition, tobacco leaf imports fluctuated over time (Figure 3-5). It appears that shortfalls in local leaf production (Figure 3-1) were covered by increasing imports. Since 1990, leaf imports have increased steadily, with the increase seen almost exclusively in countries with transnational tobacco investments.

Tobacco leaf and cigarette exports from the USSR varied between 1961 and 1990 with no clear trend but were always small, at under 5,000 metric tonnes and 5 billion units (one unit = 1 cigarette) respectively. A sudden increase in cigarette exports occurred in the mid 1990s (Figure 3-6), a trend that appears to have continued, although it must be noted, the scale of exports is a fraction that of exports. Thus the regional trade deficit in cigarettes remains substantial, although both USDA and UN FAO data suggest it has fallen since 1999.

In contrast, leaf exports have not increased overall since transition (data not shown) and the regional trade deficit in tobacco leaf has increased ten fold from 1992 to 1999.

148 Figure 3-4 Cigarette imports in the USSR/FSU, 1960 to 2001. UN data are given for the USSR configuration and USDA data for the USSR configuration until 1991 and the FSU from 1992 onwards.

140,000

120,000

USDA data USSR/FSU 100,000 UN data USSR

80,000

60,000

40,000

20,000 Millions of units (USDA data) or metric tons (UN data)

0

0 4 0 60 7 8 0 9 9 9 0 1 1962 1964 1966 1968 1 1972 1974 1976 1978 1980 1982 1 1986 1988 1990 1992 1994 1996 1998 2

Source: USDA and UN FAO agriculture and food trade database Notes: USDA 2001 data are estimates.

149 Figure 3-5 Tobacco leaf imports in the USSR/FSU, 1961- 1999.

400000

350000

USSR boundaries 300000 NIS boundaries

Countries without 250000 investment Countries with investment

200000

150000 Tobacco leaf imports (metric tons) (metric imports leaf Tobacco

100000

50000

0

9 9 7 7 67 77 85 95 961 963 9 96 971 973 9 97 981 9 98 989 991 9 99 999 1 1 1965 1 1 1 1 1975 1 1 1 1983 1 1 1 1 1993 1 1 1

Source: UN FAO agriculture and food trade database

150 Figure 3-6 Cigarette exports in the USSR/FSU, 1960-2001. UN data are given for the USSR configuration and USDA data for the USSR configuration until 1991 and the FSU from 1992 onwards.

25,000

20,000 USDA data USSR/FSU UN data USSR

15,000

10,000

5,000 Millions of units (USDA data) or metric tons (UN data) (UN tons metric or data) (USDA of units Millions

0

2 6 0 4 8 2 6 0 94 98 1960 196 1964 196 1968 197 1972 197 1976 197 1980 198 1984 198 1988 199 1992 19 1996 19 2000*

Source: USDA and UN FAO agriculture and food trade database Notes: USDA 2001 data are estimates.

Cigarette consumption Per capita cigarette consumption increased between 1960 and the mid 1970s but then stabilised for a decade until the shortfalls in production and imports led to a rapid decline until the mid-1990s (Figure 3-7). Since then consumption has increased almost exponentially and now totals almost 575 billion cigarettes per year, considerably higher than the previous peak. Over the period 1991 to 2001 per capita consumption among those aged 15+ increased by 43% in all countries combined, 56% in countries that had received tobacco industry investments compared with a 1% fall in countries that had not (Table 3-2).

151 Figure 3-7 Cigarette consumption per capita in the USSR/FSU (all ages), 1960-2001

2,500

2,000

1,500

1,000

500 No.cigarettes per of capita perconsumed year

-

4 0 2 8 60 62 6 70 72 76 78 8 8 84 86 8 92 94 96 00 9 9 9 0 1 19 19 1966 1968 19 19 1974 1 19 19 19 1 19 19 1990 19 19 19 1998 2

Source: Cigarette consumption - USDA data. Population data - UN data to 1989 taken from UN demographic yearbooks, WHO data from 1990 taken from WHO HFA database Notes: 2001 data are estimates

152 Table 3-2 Cigarette consumption per capita 15+ in the Former Soviet Union (FSU) as a whole, and in countries with and without tobacco transnational investments % increase 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001* 1991-2001 Countries without investment 2024 1757 1832 1907 1890 1940 1954 1959 1981 1960 1996 -1% Countries with investment 1712 1653 1641 1623 1805 1784 1908 1972 2184 2580 2679 56% FSU as a whole 1766 1661 1651 1647 1798 1783 1888 1944 2127 2468 2529 43% Countries without investment include: Belarus, Georgia, Moldova, Tajikistan, Turkmenistan Countries with investment include: Estonia, Kazakhstan, Latvia, Lithuania, Russia, Ukraine, Uzbekistan. Armenia, Azerbaijan and Kyrgyzstan appear only in the "FSU as a whole" group and this accounts for differences between this and the other two groups *Notes: 2001 consumption and population data are estimates. For Turkmenistan, 1999 and 2000 population data are also estimates based on 1998 data

153 DISCUSSION These data suggest that the transition to a market economy, with its accompanying liberalisation of trade and investment permitting entry of the tobacco transnationals, has had a major impact on tobacco trade and consumption in the FSU. Cigarette consumption has increased almost exponentially in line with the rapid increase in cigarette production. Moreover, these large increases in consumption have been concentrated in countries receiving tobacco industry investment. Tobacco leaf production declined, largely due to the disruption of transition, but has now started to increase, not only in traditional producing areas but also in Uzbekistan and Kazakhstan, following British American Tobacco and Philip Morris investments. Nevertheless, since transition the trade deficit in tobacco leaf has increased ten-fold belying TTC claims that investment would improve trade figures as explored further in Chatper 7. Cigarette imports increased only temporarily and in countries receiving industry investments, seemingly until output from the updated local production facilities had reached a sufficient level, while exports have seen a continued but smaller rise insufficient to result in a trade surplus.

Before considering the results in any detail, it is necessary to consider data accuracy. There are three main concerns in this area – data collection systems, smuggling and illegal production. With independence, each country had to establish new data collection systems and this caused difficulties particularly in the early 1990s. Thus while the Statistical Committee of the Commonwealth of Independent States publishes trade statistics, they do not cover all years or all countries, and the definitions of tobacco related categories are inconsistent.313 Similarly, the United National Statistic Division Comtrade database does not contain data for some countries in the region such as Uzbekistan. The USDA attempts to overcome these problems by using best estimates in the absence of credible data.

Problems with data appear mainly to have affected the smaller central Asian states which contribute less to the regional total. In addition, the gaps are predominantly in export data for the early 1990s which, given the small scale of exports relative to production or imports, will have relatively little impact on final consumption figures.

154 Import and export data include only officially traded cigarettes and are therefore problematic given that smuggling was and is a major issue in the region (see Chapter 6).211 As much of the smuggling, particularly in more recent years, is likely to occur between countries within the region, this problem is overcome to some extent by considering the FSU as a whole. Based on the fact that cigarette exports far outweigh imports, it is estimated that approximately one third of global cigarette exports are smuggled. 314 The data presented here show the opposite occurring in this region. Nevertheless, the most likely impact of smuggling on the data presented in this chapter is to underestimate imports to and hence consumption in countries without substantial TTC investments, whilst also perhaps underestimating exports from countries that have received investments. This will exaggerate the difference in the changes in consumption between countries with and without tobacco industry investments and may account for part of these differences. Overall, the three issues are likely to underestimate consumption in the immediate post-transition period when data problems, illegal production and smuggling (used as an industry market entry strategy) were greatest.315

Whilst not wishing to overlook these serious concerns, the data presented here are the most comprehensive available and, despite their weaknesses, allow a preliminary assessment of an important issue. Although data sources often overlap, wherever possible, data from at least two sources was obtained and compared. FAO data were used for tobacco leaf production as they were more consistent with industry data on production levels in the post transition period308,83 and because the trade data have the advantage of being presented in metric tons across the whole time period. UN and USDA data on imports, exports and production differed somewhat more in the post- transition period, particularly in the early 1990s, although overall trends were similar. USDA data were more complete and believed to be more accurate for consumption (and for the underlying import, export and production data). They were, for example, more consistent with what was reported in the publication World Tobacco Trends316 and with ERC data211 and had the additional advantage of being more up to date. Other sources, including the World Tobacco File317 did not have data for all countries in the region and could not therefore be used.

155 Attempts to validate the findings also suggest they are reasonably robust. For most countries, the estimates of cigarette consumption per capita were very similar to those provided by ERC, which also found that between 1990 and 2000 consumption increased by 57.3% in Russia.211 In addition, existing survey data, although limited, also suggest that smoking prevalence has been rising particularly among young women (see Chapter 1, Sections 3.3 and 3.4),67,129,318 a finding supported by this thesis (see Chapter 4). Production figures were broadly consistent with the production capacity figures presented in Chapter 2. These suggest that production capacity in factories with TTC investments totalled 416 billion cigarettes compared with a total production figure in 2000 of 506 billion, which, suggests that 90 billion are produced in locally or state owned factories.

This current level of consumption (2529 cigarettes per capita in 2001) is high by international standards although similar to levels seen in much of central and eastern Europe.117 The increase in consumption has been far greater in countries that have received major tobacco industry investments (56%) than in countries that have not (- 1%). While some of the overall increase is due to the artificially low consumption levels seen around independence and some of the differential increase in countries receiving investments is due to undocumented smuggling from these countries in the latter half of the decade (or smuggling to these countries in the first half), it is clear that consumption has now increased well above its previous peak in the mid 1980s. The increase in consumption is particularly notable for two reasons. Firstly, the tobacco epidemic in the FSU, at least amongst men, has been established for some time. Although historical data on smoking habits are scarce, contemporary studies asking about ever smoking, combined with data on lung cancer mortality145,195 suggest that male smoking must have become widespread during the first half of the twentieth century, probably contemporaneously with the establishment of the habit in the US or UK. In addition, the region’s first cigarette factories and brands were established in the 1850s and 1860s.63,224 Although Soviet women did not smoke in large numbers until recently, the classic description of the progress of the tobacco epidemic would suggest that consumption should now be steady or declining as appeared to be the case in the 1970s to 1980s, not increasing to the extent indicated here. Second, after independence, most countries

156 experienced sustained economic recessions, with marked increases in poverty, which would be expected to reduce rather than increase consumption.

As noted earlier, trade liberalisation can work in several ways to increase consumption. Although elucidating the precise mechanisms in this instance is impossible because of the absence of detailed data on, for example, price, compounded by the extreme financial volatility during this period, it appears that many factors seen elsewhere were also in operation here, albeit with some minor differences. An increased supply of cigarettes was seen but occurred through increased production rather than imports. Even where companies planned to or successfully established monopolies (Chapter 2)319 and then exerted pressure on governments to close the market to outside competition through both tariff and non-tariff barriers,320,321 competition was more intense than in the Soviet era. Moreover, industry documents suggest that in terms of marketing, such markets would be treated as though they were competitive (for more on this see Chapters 8, 9 and 11). 319 Thus advertising increased virtually everywhere as outlined in Chapter 2, and even where the TTCs had manufacturing monopolies as will be detailed in Chapter 11. As will be outlined in more detail in Chapters 6, 8 and 11, tobacco industry documents indicate that young people, women, opinion leaders and urban residents were specifically targeted and the allure of western products was used to attract smokers.319,322,323,324 This targeted advertising combined with the increased production of filter brands, the introduction of milder brands and brands targeted specifically to women, to a market previously dominated by coarse filterless or papirossy cigarettes must have encouraged new smokers (particularly women) to take up the habit as the TTCs predicted;319 an issue examined further in Chapter 4, which presents the survey results.

But why then was the increase in consumption in the FSU (approximately 40%) so much greater than in Asia (10-20%)? One possible factor is data artefact, in particular the artificially low consumption level at the end of the 1980s. Another is the absence of effective tobacco control policies, or even organised tobacco control groups that might have counterbalanced industry pressure. This contrasts with the position in Thailand and Taiwan where the entry of the TTCs acted as a stimulus to tobacco control. It is argued here that a major factor was the enormous economic and political leverage of the

157 tobacco industry on account of their major contribution to FDI in the recipient countries (see Chapter 2). The relatively greater increase in consumption is also consistent with Shepherd’s assertion (see Chapter 1) that demand creation cannot be fully exploited through imports. By buying established factories and brands (in the FSU often bought separately to the factories) the TTCs were able to benefit from established distribution systems and brand loyalty. By then manipulating established brands and introducing new western brands and basing marketing efforts on both, the TTCs may have been more easily able to increase demand, as Weissman has previously suggested may occur.10 It is also likely that efforts to stimulate demand succeeded because of the vulnerability of the population in a time of rapid transition and great uncertainty. My previous research on factors influencing smoking behaviour in Ukraine highlighted the role of deterioration in social position (a proxy for of the stress of changes associated with transition), as well as unemployment and poverty as important determinants of current smoking.67

Over the period 1991 to 2000, annual cigarette production increased by over 200 billion, a 76% increase. Yet, despite the increase in production, exports have only increased by a fraction of this amount to a total of just over 20 billion for the region. Meanwhile, imports, which rose initially, have declined to approximately 50-60 billion. Overall, therefore, despite significant investment in the region’s tobacco industry, the overall trade balance in cigarettes remains negative although it does appear to have improved recently. Moreover, the rapid increase in tobacco leaf imports, seen almost exclusively in countries receiving tobacco industry investment, has led to a rapid increase in the trade deficit. The shift in consumer preferences towards new blended cigarette varieties, coupled with the increase in consumption, has led to a decline in the proportion of tobacco imported from former Soviet republics in favour of imports from, for example, India, Greece, Turkey, Italy, Spain, Zimbabwe and Brazil.325,326

Ideally, further work would seek to verify these findings on a country by country basis using more detailed econometric analysis to control for changes in incomes, price and advertising over this period. I considered doing so here but it rapidly became clear that many factors (not only affecting data on tobacco) that arose as countries struggled to establish data systems, tackle the informal economy, introduce new currencies, re-

158 denominate old ones, deal with hyper-inflation, build state structures and, in some cases, define national frontiers following the outbreak of hostilities, made it impossible to obtain sufficient valid and meaningful data, particularly financial data, in which one could be confident. Nevertheless, I conclude tentatively that similar to trade liberalisation, liberalisation of inward investment leads to an increase in cigarette consumption. Although liberalisation has led to the investment of much needed capital, as I shall explore further in Chapters 7, 8, 9 and 12, the investments were lower than anticipated, and no other benefits have accrued from tobacco industry investments. Trade deficits initially increased and, although now stabilised, have yet to decline and leaf deficits are likely to increase. Profits from tobacco sales will accrue to investors outside the region, while the considerable costs of long term health consequences will be borne by host countries with already high premature mortality rates.

Despite its own evidence on the impact of trade liberalisation on tobacco consumption, the World Bank has previously suggested that trade restrictions would be potentially counterproductive, and recommends that supply side measures be restricted to the control of smuggling. 327 Others however argue that international treaties that have liberalised trade should develop specific rules to govern tobacco as they have done with other harmful products such as weapons and hazardous waste – products that kill far fewer people.145,328 In the meanwhile, a basic first step would be to protect markets before their opening through ensuring the presence of comprehensive tobacco control programmes with comprehensive advertising bans and effective taxation policies as absolute prerequisites. The IMF which may pressure for industry privatisation (as it did for example in Moldova10,237) and the World Bank have a particular responsibility in this regard. Unlike the World Bank which recognises the economic consequences of tobacco use and poor health, this may require a major volte-face by the IMF.

The next chapter builds on the evidence presented here by examining the prevalence of smoking and its socio-demographic correlates in eight former Soviet republics.

159

CHAPTER 4 Prevalence and determinants of smoking in eight countries of the former Soviet Union

BACKGROUND Despite the deplorably high levels of tobacco-related mortality outlined in Chapter 1, relatively little is known about smoking prevalence in the former Soviet Union and even less about the determinants of smoking behaviour. Virtually no recent or reliable data exist for the central Asian countries (Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan and Uzbekistan).329,67 Recent surveys conducted in Georgia are limited to the capital Tbilisi.124,125 Data from elsewhere in the Caucasus (Armenia, Azerbaijan) are scarce118 and historical figures68 inconsistent with later findings, leading authors to rely on anecdotal reports of smoking rates.330

Historicall67 and more recent data, which come largely from Russia,128 the Baltic States131 and my own work in Ukraine129 and Belarus130 show, perhaps unsurprisingly, given the mortality figures described above, that smoking rates in men are high at between 45% and 60%; rates are far lower in women, varying from 1% to 20%.329 The higher rates previously seen in Estonian women are now being matched by those in the other Baltic States329,131,127 and by women in urban areas elsewhere.128,129 These same surveys128,129,130,131 suggest that gender, age, urbanisation and socio-economic status influence the likelihood of smoking, although a number of between-country variations were reported. Unfortunately however, other than the Baltic states, few countries collect data using consistent data collection tools thereby precluding accurate between-country comparisons. Thus, part of the variation observed between studies could be due to differences in methodology or choice of diverse markers of socio-economic status.

These issues underlie the need in the FSU for comparable and accurate data on smoking prevalence and its determinants, widely recognised as a pre-requisite for the development of effective public health policies.12,11,112 This need is made more urgent by the profound changes experienced in the FSU’s recent economic transition and more specifically by the changes to its tobacco industry, as described in Chapter 2.242

This chapter aims to provide timely, comparable data on smoking behaviour and its demographic and socio-economic determinants in eight of the FSU countries.

160

METHODS The LLH Project The Living Conditions, Lifestyles and Health (LLH) Project in which the European Centre on Health of Societies in Transition (ECOHOST) is a partner, is a European Community funded study of health in eight of the former Soviet Union countries. The project specifically explores how health in the populations of Armenia, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Moldova, Russia and Ukraine is affected by life- style including smoking habits. It is overseen by an Executive Committee on which representatives of the study partners sit.

As part of the study a survey of approximately 18500 respondents, was conducted in the autumn of 2001. The survey was approved by the ethics committee of the London School of Hygiene and Tropical Medicine. Verbal informed consent was obtained from all study participants at the beginning of the interviews.

Study population and sampling procedures Quantitative cross-sectional surveys on living conditions, lifestyle and health were conducted in the eight FSU countries, by organisations with expertise in survey research using standardised methods. 331 Each survey sought to include representative samples of the national adult population aged 18 years and over, although a few small regions had to be excluded because of geographic inaccessibility, socio-political situation or ongoing military actions: Abkhazhia and Osetia in Georgia (3% of the national population), the Trans-Dniester region and municipality of Bender in Moldova (approximately 15% of the population) and the Chechen and Ingush Republics and the autonomous districts located in the far north of the Russian Federation (2% of the population).

Samples were selected using multi-stage random sampling with stratification (except Kyrgyzstan) by region and area (rural/urban settlement types). Within each primary sampling unit, households were selected using standardised random route procedures, except in Armenia where random sampling from household lists was used. Within each household the adult with the nearest birthday was selected for interview. Individuals in the military and prisoners were excluded from the samples. Other exclusion criteria (with slight variations among countries) included being mentally disabled, institutionalised, hospitalised or homeless, or suffering from heavy alcoholic intoxication.

161

If after three visits (different days/times) there was no one at home, the next household on the route was selected. In addition, some countries allowed for substitution when the household was not used for residence, the building was ruined (disaster zone), or the selected individual was not in the country at the time of the survey. Some pre-specified quota control was used in Belarus, Kazakhstan, Moldova, and Ukraine (combination of region, area, gender, age, and/or education level). In Georgia and Russia there was no quota control during field work but a sampling repair procedure based on area, gender, age and education level. Interviews were conducted primarily in the language of the country and in Russian.

It was decided to include at least 2000 respondents in each country, but to boost this number to 4000 in the Russian Federation and to 2500 in Ukraine to reflect the larger and more regionally diverse populations in those countries. It is expected that a sample size of 2000 would give reliable estimates of proportions that represent 3% or more of the population at the national level with a precision level of 0.75%.

Questionnaire design The first draft of the questionnaire was developed in consultation with country representatives from pre-existing surveys conducted in other transition countries and from the New Russia Barometer surveys332 adjusted to the national context. I was responsible for developing the questions on tobacco use, which drew on recent developments in question design and my previous work in the region.129,130 They went beyond conventional questions on smoking frequency to explore circumstances in which smoking took place and to examine health beliefs related to smoking. However, due to the broad topic areas covered by the survey, question numbers had to be limited.

The questionnaire was developed in English, translated into appropriate national languages, back translated to check consistency, and piloted in each country. The pilot showed the questionnaire was too long for the average respondent to be willing to answer. The Executive Committee then took responsibility for cutting the length of the questionnaire at a meeting in Ukraine in September, 2001.

Questionnaire administration Questionnaires were administered by trained interviewers using face-to-face interviews conducted in respondents’ homes. Quality control procedures were part of fieldwork and generally included re-interviews (home visits or telephone calls) with the

162

respondents to assess the work of both the interviewers and the interviewers’ supervisors.

Statistical analyses Surveys were coded into SPSS (SPSS Inc.). Data were merged and converted into STATA version 6 (Stata Corporation, College Station, Tex) for statistical analysis. The continuous variables “age of first smoking” and “smoking duration” were transformed before analyses using log normal transformations to reduce the level of skewness of their distribution but returned to their original units in the tables of results. Current smokers were defined as respondents reporting currently smoking at least one cigarette per day.

Analysis of smoking prevalence rates Age and sex-specific smoking prevalence was calculated for each country. Among current smokers, the age of first smoking and number of cigarettes smoked was examined. Levels of nicotine dependence were assessed by identifying the proportion of current smokers who consume over 20 cigarettes per day and smoke within an hour of waking. This is equivalent to a score of 3 or more on the abbreviated Fagerstrom dependency scale333 , 334 and indicates moderate (score 3-4) to severe dependency (score >5) (due to the way in which data were collected it was impossible to break the score down further than this). Within each country, gender differences in smoking habits were assessed using chi-squared tests and two-sample t-tests; variations by age group were estimated using logistic regression analyses taking 18-29 year olds as the reference category. Between country comparisons in the likelihood of smoking were made using logistic regression taking Russia as the baseline, while comparisons in the geometric mean age of first smoking were made using ANOVA combined with Bonferroni multiple comparison tests. To allow for the large number of comparisons being made, 99% confidence intervals were used and significance was taken as <0.01.

Analysis of smoking determinants Selected demographic and socio-economic correlates of smoking examined included age (stratified by 10 years age group), area of residence, marital status, being of Muslim religion, education level, economic situation of the family this year, and lack of social support. Social support was derived from five yes/no questions asking

163

whether the respondents had anyone: 1) you can really count on to listen to you when you need to talk; 2) you can really count on to help you out in a crisis; 3) you can totally be yourself with; 4) you feel appreciates you as a person; and 5) you can really count on to comfort you when you are very upset. Respondents could achieve a maximum score of 5, categorised as follows: 1) score of 5 = no lack of social support (69% of respondents); 2) score of 4 or 3 = some lack (12% and 7% of respondents respectively); and 3) 2 or 1 or 0 = severe lack (4%, 2%, 6% of respondents respectively).

The association of smoking with each demographic and socioeconomic factor was estimated using logistic regression analyses, stratifying by gender. In a first step, age- adjusted odds ratios associated with different levels of each factor were calculated. Then, multiple logistic regression analyses were used, with simultaneous adjustment for all the factors in the regression model. Multiple regression analyses were also conducted using data from all countries, adjusting for country of residence. Once again, because several statistical tests were performed, statistical significance was taken as p<0.01.

RESULTS Response Rates 18,428 individuals were surveyed. Response rates varied between 71% and 88% among countries (calculated using the total number of households for which an eligible person could be identified). Item non-response rates were very low, for example 0.03% for current smoking, 0.5% for education level.

Sample characteristics and representativeness The samples clearly reflect the diversity of the region and are broadly representative of the populations they denote (Table 4-1). Comparisons with official data may be limited by the failure of some country data to fully capture post-transition migration and other shortfalls in routine data335 but suggest a slight under-representation of men in Armenia and Ukraine, of the urban population in Armenia and the rural population in Kyrgyzstan. Age group comparisons for those in the sample aged 20 plus suggest there is a tendency for the oldest age group to be over-represented at the expense of the youngest particularly in Armenia, Moldova and Ukraine.

164

Table 4-1 Characteristics of samples and countries

Characteristic AM BY GE KZ KG MD RU UA Sample data Response rate (%) 88 73 88 82 71 81 73 76 Gender - % males in whole sample (18+) 40.3 44.1 45.7 44.4 45.0 45.1 43.5 38.8 (% males aged 20+) (40.7) (43.9) (45.6) (44.1) (45.6) (44.9) (43.2) (38.6) (n) 2000 2000 2022 2000 2000 2000 4006 2400 Age group (% of total aged 20+) 20-29 years 15.4 16.9 13.9 21.9 26.7 14.5 16.5 14.6 30-39 years 21.6 19.2 20.3 25.8 26.0 20.1 19.3 16.4 40-49 years 24.0 21.6 21.9 21.5 21.4 23.1 20.9 17.9 50-59 years 11.1 14.5 16.3 12.0 10.1 16.4 15.4 15.5 60+ years 28.0 27.9 27.6 18.8 15.9 26.0 27.9 35.5 (n) aged 20+ 1940 1922 1975 1890 1899 1945 3828 2324 (n) aged 18-19 60 78 47 110 101 55 178 76 Administrative classification of interview place (%) State/regional capital 44.0 33.9 41.4 27.0 27.5 30.4 35.7 31.5 Other city/small town 17.0 34.8 15.6 25.4 13.5 11.6 37.1 36.4 Village 39.0 31.4 43.0 47.6 59.0 58.1 27.3 32.1 (n) 2000 2000 2022 1850 2000 2000 4006 2400 Reported nationality (%) b Nationality of country 97.3 80.1 90.2 36.3 68.6 76.7 82.4 77.7 Russian 0.8 12.1 1.3 41.5 18.0 7.7 - 16.5 Other 1.9 7.8 8.5 22.1 13.5 15.7 17.6 5.8 (n) 2000 1979 2021 1979 1997 1980 3967 2371 Education (%) Secondary education or less 49.1 49.4 33.8 35.7 48.3 52.2 43.2 44.2 Secondary vocational or some higher education 30.4 34.2 32.7 43.5 32.7 32.7 35.7 36.1 Completed higher education 20.5 16.4 33.6 20.8 19.0 15.2 21.1 19.7 (n) 1996 1984 1996 1995 1996 1984 4004 2381 c Country data Mid-year popn ('000, in 2001) 3788 9971 5238 14821 4927 4254 144387 49111 GNP per capita (US$, in 2001) 560 1190 620 1360 280 380 1750 720 % males in popn aged 20+ years (in 2000) 47.5 45.4 46.4 46.6 47.9 46.3 45.3 44.8 % urban popn (in 2001) 67.3 69.6 56.5 55.9 34.4 41.7 72.9 68.0 Age group (% of total age 20+) 20-29 years 23.2 19.3 20.6 26.0 30.5 23.1 19.6 19.4 30-39 years 24.2 20.3 21.1 23.7 24.7 20.3 19.6 19.0 40-49 years 22.5 21.5 19.5 21.4 19.6 22.7 22.4 19.8 50-59 years 10.3 12.6 12.7 10.9 9.0 13.6 13.3 14.2 60+ years 19.7 26.4 26.2 18.0 16.2 20.3 25.1 27.6 d % unemployment rate 11.7 2.3 11.1 2.9 3.2 2.0 13.4 5.8 Tobacco industry - state owned (SO) or privatised (P) P SO P P P SO P P Foreign direct investment in tobacco industry to end of 2000 (US$millions)e 8 0 0 440 n/a 0 1719 152.9 FDI in tobacco industry per capita x 1000 e 0.002 0.000 0.000 0.030 n/a 0.000 0.012 0.003 a Average assuming the same number of respondents in each country. b Means Armenian in Armenia, Belarussian in Belarus, Georgian in Georgia, Kazakh in Kazakhstan, Kirghiz in Kirgyzstan, Moldovian/Romanian in Moldova, Russian in Russia and Ukrainian in Ukraine. c Data sources: 1) European Health for All Database, January 2003; 2) Population Division of the Dpt of Economic and Social Affairs of the UN Secretariat, World Population Prospects: The 2002 Revision and World Urbanization Prospects: The 2001 Revision, http://esa.un.org/unpp; d In 1999 for Russia, 2000 for Armenia and Ukraine, and 2001 for the other countries. e Data from Chapter 2 NB these are minimum investment figures

165

Smoking prevalence Rates of male smoking were high. In many of the countries surveyed almost 80% of men have ever smoked (Table 4-2, Figure 4-1). The prevalence of current smoking is lowest in Moldova (43.3%) and Kyrgyzstan (51.0%) and highest in Kazakhstan (65.3%), Armenia (61.8%), Russia (60.4%) and Belarus (56.1%). Current smoking rates in Russia could not be distinguished from those in Kazakhstan, Armenia or Belarus but were significantly higher than in Moldova, Kyrgyzstan, Ukraine and Georgia (p<0.01, data not shown).

Rates in women were far lower (p-value for gender comparisons <0.001 in all countries) and somewhat more variable, ranging from 2.4% to 15.5% with the lowest rates seen in Armenia, Moldova and Kyrgyzstan and the highest in Russia, Belarus and Ukraine. Current smoking among women in Russia was significantly more prevalent than in all other countries (p<0.01) although adjusting for age removed the difference between Russia and Belarus (data not shown).

The relationship between smoking and age varied by gender. In men, with the exception of Moldova, current smoking prevalence varied little between the ages of 18 and 59 but then declined more markedly in the over 60s (Table 4-2, Figure 4-2). This decline with age was accounted for by an increase in the proportion of ex- and never-smokers among older groups. Among women, the overall trend was for both current and ex-smoking to decrease with increasing age, with very low smoking rates observed in the oldest age group (ever smoking rates varied from 0.8-3.9% in those aged 60 plus). However, closer inspection of the data suggests that the countries could be divided into two groups. In the first (Russia, Belarus, Ukraine, Kazakhstan), rates of current and ever smoking imply that initiation of smoking has increased rapidly between generations and especially in the youngest age group (Figure 4-2- Figure 4-3). In the second group (Armenia, Georgia, Kyrgyzstan and Moldova) the age trends were less obvious and were non-significant (except when comparing the oldest and youngest age groups in Moldova).

166

Table 4-2 Smoking prevalence by country, gender and age group

Males Females Sex differences in b (n) All age groups 18-29 30-39 40-49 50-59 60+ (n) All age groups 18-29 30-39 40-49 50-59 60+ current smoking % 99% CI % % % % % % 99% CI % % % % % ARMENIA (AM) Current smoker 498 61.8 56.2 67.4 62.5 76.8 68.3 67.1 44.4 28 2.4 -5.0 9.7 0.9 3.1 3.9 2.9 1.0 p=<0.001 Ex-smoker 120 14.9 6.5 23.3 8.3 5.5 14.2 17.1 25.1 7 0.6 -6.8 8.0 0.9 0.4 0.4 0.7 0.7 Never smoker 188 23.3 15.4 31.3 29.2 17.7 17.5 15.8 30.5 1159 97.1 95.8 98.3 98.1 96.5 95.7 96.4 98.4 Odds of current smoking 1.00 1.98 1.29 1.22 0.48 1.00 3.43 4.3 3.15 1.05 p-value 0.006 0.272 0.499 0.001 0.121 0.059 0.19 0.952 BELARUS (BY) Current smoker 495 56.1 50.4 61.9 58.2 65.3 59.8 60.2 40.3 135 12.1 4.9 19.3 30.4 18.5 12.7 3.1 0.9 p=<0.001 Ex-smoker 125 14.2 6.1 22.2 9.2 12.1 12.9 11.0 23.9 60 5.4 -2.1 12.9 13.5 7.7 2.3 4.4 1.5 Never smoker 262 29.7 22.4 37.0 32.7 22.5 27.3 28.8 35.8 922 82.5 79.3 85.8 56.0 73.9 85.1 92.5 97.6 Odds of current smoking 1.00 1.35 1.07 1.08 0.49 1.00 0.52 0.33 0.07 0.02 p-value 0.159 0.743 0.726 <0.001 <0.001 <0.001 <0.001 <0.001 GEORGIA (GE) Current smoker 491 53.3 47.4 59.1 62.8 64.8 61.5 50.7 33.9 69 6.3 -1.2 13.9 5.8 11.6 7.8 3.4 3.4 p=<0.001 Ex-smoker 71 7.7 -0.5 15.9 2.0 4.4 4.5 10.4 14.5 10 0.9 -6.8 8.7 2.3 1.4 1.3 0.0 0.0 Never smoker 360 39.1 32.4 45.7 35.1 30.8 34.0 38.9 51.6 1012 92.8 90.7 94.9 91.9 87.0 90.9 96.6 96.6 Odds of current smoking 1.00 1.09 0.94 0.61 0.30 1.00 2.13 1.38 0.58 0.57 p-value 0.707 0.799 0.037 <0.001 0.051 0.426 0.295 0.219 KAZAKHSTAN (KZ) Current smoker 579 65.3 60.2 70.4 66.0 72.7 65.9 64.2 50.0 103 9.3 1.9 16.6 16.1 10.9 11.2 3.4 0.4 p= <0.001 Ex-smoker 119 13.4 5.4 21.5 7.6 9.5 16.2 18.4 24.2 48 4.3 -3.2 11.9 5.8 7.8 4.3 0.9 0.4 Never smoker 189 21.3 13.6 29.0 26.4 17.8 17.9 17.4 25.8 962 86.4 83.6 89.3 78.1 81.3 84.6 95.8 99.1 Odds of current smoking 1.00 1.37 1.00 0.92 0.52 1.00 0.64 0.66 0.18 0.02 p-value 0.111 0.982 0.744 0.003 0.087 0.113 0.002 <0.001 KYRGYZSTAN (KG) Current smoker 457 51.0 44.9 57.0 56.2 60.4 49.8 50.0 25.0 49 4.5 -3.1 12.1 4.2 5.4 6.0 4.9 1.7 p= <0.001 Ex-smoker 79 8.8 0.6 17.0 4.9 5.5 8.3 6.8 25.8 22 2.0 -5.7 9.7 2.7 1.8 2.5 0.0 1.7 Never smoker 361 40.3 33.6 46.9 39.0 34.1 42.0 43.2 49.2 1022 93.5 91.5 95.5 93.1 92.8 91.5 95.1 96.7 Odds of current smoking 1.00 1.19 0.77 0.78 0.26 1.00 1.31 1.45 1.18 0.39 p-value 0.353 0.166 0.313 <0.001 0.474 0.357 0.759 0.140 MOLDOVA (MD) Current smoker 390 43.3 36.8 49.8 62.6 52.4 44.9 38.3 24.7 43 3.9 -3.7 11.5 6.0 7.7 2.5 3.6 1.1 p= <0.001 Ex-smoker 125 13.9 5.9 21.8 6.5 10.1 13.2 16.9 20.6 13 1.2 -6.5 8.9 3.9 1.4 0.8 0.6 0.0 Never smoker 386 42.8 36.4 49.3 31.0 37.5 42.0 44.8 54.8 1043 94.9 93.1 96.7 90.1 91.0 96.7 95.8 99.0 Odds of current smoking 1.00 0.66 0.49 0.37 0.20 1.00 1.29 0.39 0.59 0.16 p-value 0.065 0.001 <0.001 <0.001 0.526 0.070 0.304 0.006 RUSSIA (RU) Current smoker 1052 60.4 56.5 64.3 66.4 69.7 68.4 59.9 42.3 348 15.5 10.5 20.5 30.6 23.8 13.1 13.0 2.5 p= <0.001 Ex-smoker 308 17.7 12.1 23.3 10.1 13.6 11.6 18.5 31.9 135 6.0 0.7 11.3 11.2 7.8 6.4 5.1 1.4 Never smoker 381 21.9 16.4 27.3 23.5 16.8 19.9 21.6 25.9 1768 78.5 76.0 81.1 58.2 68.5 80.5 81.9 96.0 Odds of current smoking 1.00 1.16 1.10 0.75 0.37 1.00 0.71 0.34 0.34 0.06 0.360 0.558 0.910 <0.001 0.025 <0.001 <0.001 <0.001 p-value UKRAINE (UA) Current smoker 488 52.5 46.7 58.4 61.9 65.2 56.5 59.5 35.7 162 11.1 4.7 17.4 32.9 15.3 9.2 8.7 1.0 p = <0.001 Ex-smoker 157 16.9 9.2 24.6 11.4 6.5 14.1 13.0 28.3 40 2.7 -3.9 9.4 5.4 3.7 3.9 2.2 0.8 Never smoker 284 30.6 23.5 37.6 26.7 28.3 29.4 27.5 36.0 1261 86.2 83.7 88.7 61.7 81.0 86.9 89.1 98.3 Odds of current smoking 1.00 1.15 0.80 0.90 0.34 1.00 0.37 0.21 0.19 0.02 p-value 0.549 0.297 0.671 <0.001 <0.001 <0.001 <0.001 <0.001 ALLa Current smoker 4417 55.5 53.5 57.4 62.1 65.9 59.4 56.2 37.0 846 8.1 5.7 10.5 15.9 12.0 8.3 5.4 1.5 Ex-smoker 1070 13.4 10.7 16.1 7.5 8.4 11.9 14.0 24.3 301 2.9 0.4 5.4 5.7 4.0 2.7 1.7 0.8 Never smoker 2479 31.1 28.7 33.5 30.4 25.7 28.7 29.8 38.7 9274 89.0 88.2 89.8 78.4 84.0 89.0 92.9 97.7 Odds of current smoking 1.00 1.19 0.91 0.76 0.36 1.00 0.70 0.45 0.34 0.08 p-value 0.018 0.155 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 Between country differences in <0.001 0.195 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.032 current smoking b a b Average assuming the same number of respondents in each country. using chi squared test on binary variable current versus never and ex-smokers

167

Prevalence (%)

80

70

60

50

Current smoking Females 40 Current smoking Males Ever smoking Females 30 Ever smoking Males

20

10

Ever smoking Males 0 Ever smoking Females MD KG Current smoking Males UA GE BY Current smoking Females RU AR KZ

Figure 4-1 Prevalence of current and ever smoking by gender

168

Figure 4-2 Current male smoking prevalence by age group

90 80 Armenia 70 Belarus 60 Georgia 50 Kazakhstan 40 Kyrgyzstan Moldova 30 Russia 20 Ukraine 10

0 18-29 30-39 40-49 50-59 60+

Age group

Figure 4-3 Current female smoking prevalence by age group

35

30 Armenia

25 Belarus

Georgia 20 Kazakhstan 15 Kyrgyzstan Moldova

10 Russia

Ukraine 5

0

18-29 30-39 40-49 50-59 60+

Age group

169

Table 4-3 Characteristics of smoking behaviour amongst current smokers AM BY GE KZ KG MO RU UA Alla Between country comparisonsb %%%%%%%%% Age when started smoking (years) Males Mean age 18.5 17.4 18.2 17.6 19.1 18.2 17.0 17.2 17.9 Geometric mean age 17.8 16.6 17.7 17.1 18.6 17.6 16.2 16.2 17.2 p<0.001 <16 22.2 32.8 18.0 27.9 14.7 22.8 36.4 35.2 26.2 p<0.001 16-20 56.8 54.2 66.0 57.0 61.8 59.9 49.8 48.5 56.7 >20 21.0 13.0 16.0 15.1 23.5 17.3 13.9 16.3 17.0 (n) 447 430 400 502 408 347 993 435 3962 Females Mean age 28.0 18.9 22.7 20.7 21.5 23.0 20.9 21.2 22.1 p<0.001 Geometric mean age 27.0 18.5 21.3 19.9 20.7 21.5 19.8 19.9 21.1 p<0.001 <16 0.0 20.0 18.5 15.4 12.5 22.9 13.1 15.1 14.7 16-20 14.3 56.7 38.5 50.6 43.8 22.9 52.6 57.2 42.1 >20 85.7 23.3 43.1 34.1 43.8 54.3 34.4 27.6 43.3 (n) 28 120 65 91 48 35 329 152 868 Between gender comparison in geometric mean age c <0.001 0.002 <0.001 <0.001 0.002 <0.001 <0.001 <0.001 Number of cigarettes smoked daily Males One or two 1.8 3.4 1.9 4.5 15.4 8.2 2.4 4.6 5.3 p<0.001 Up to 10 18.7 32.3 12.7 30.9 50.1 43.3 24.6 25.4 29.8 10-20 51.4 50.5 63.3 48.0 28.7 37.4 52.2 53.5 48.1 >20 28.1 13.7 22.2 16.6 5.8 11.0 20.8 16.5 16.9 odds ratio for likelihood of smoking >20 /day 1.4875 0.6057 1.0853 0.756 0.2338 0.4713 1.00 0.753 p-value 0.002 0.001 0.539 0.038 <0.001 <0.001 0.049 (n) 498 495 482 579 449 390 1052 484 4429 Females One or two 32.1 23.7 11.9 19.4 36.2 37.2 18.7 22.2 25.2 p=0.065 Up to 10 28.6 48.9 29.9 53.4 46.8 41.9 56.6 45.7 44.0 10-20 32.1 25.2 46.3 23.3 17.0 18.6 19.8 26.5 26.1 >20 7.1 2.2 11.9 3.9 0.0 2.3 4.9 5.6 4.7 odds ratio for likelihood of smoking >20 /day 1.50 0.44 2.64 0.79 xx 0.46 1.00 1.15 p-value 0.602 0.199 0.032 0.672 xx 0.461 0.749 (n) 28 135 67 103 47 43 348 162 933 Between gender comparison of % smoking >20 / day d 0.015 0.000 0.053 0.001 0.090 0.073 <0.001 <0.001 Table 3 (cont'd) Time when usually smoke the first cigarette Males First 30 min after getting up 63.5 47.9 52.9 42.8 39.0 44.1 56.5 55.8 50.3 p<0.001 First hour after getting up 24.9 40.4 34.0 46.6 39.4 38.2 34.3 33.3 36.4 Before mid-day meal 4.6 6.9 5.0 5.0 7.1 6.7 4.7 6.0 5.7 After day meal or in the evening 7.0 4.9 8.1 5.5 14.5 11.0 4.6 5.0 7.6 Odds ratio for likelihood of smoking in first hour 0.77 0.77 0.67 0.86 0.37 0.47 1.00 0.83 p-value 0.140 0.129 0.021 0.394 <0.001 <0.001 0.292 (n) 498 495 480 579 449 390 1051 484 4426 Females First 30 min after getting up 50.0 31.9 44.6 35.0 27.7 14.3 33.7 27.8 33.1 p=0.278 First hour after getting up 14.3 28.9 30.8 27.2 31.9 38.1 32.0 32.1 29.4 Before mid-day meal 3.6 19.3 12.3 13.6 12.8 11.9 13.5 17.3 13 After day meal or in the evening 32.1 20.0 12.3 24.3 27.7 35.7 20.8 22.8 24.5 Odds ratio for likelihood of smoking in first hour 0.94 0.81 1.60 0.86 0.77 0.57 1.00 0.78 p-value 0.879 0.307 0.129 0.505 0.409 0.092 0.203 (n) 28 135 65 103 47 42 347 162 929 Between gender comparison in % smoking in first hour d <0.001 <0.001 0.014 <0.001 0.004 <0.001 <0.001 <0.001 Moderate to heavy nicotine dependence (>20 cigarettes/day and smoking within first hour of getting up) Males 26.9 13.7 21.4 16.6 5.6 10.5 20.6 16.2 16.4 0.000 Odds ratio for likelihood of having moderate to severe dependency 1.42 0.62 1.05 0.77 0.23 0.45 1.00 0.74 0.8 p-value 0.005 0.093 0.142 0.104 0.000 0.000 0.042 0.0 (n) 498 495 477 579 449 390 1051 483 4422 Females 7.1 2.2 10.8 3.9 0.0 1.0 17.0 9.0 6.4 0.139 Odds ratio for likelihood of having moderate to severe dependency 1.49 0.44 2.34 0.78 xx 0.47 1.00 1.14 1.0 p-value 0.605 0.197 0.071 0.669 xx 0.473 0.754 0.3 (n) 28 135 65 103 47 42 347 162 929 Between gender comparison in dependencyd 0.020 <0.001 0.045 0.001 0.097 0.091 <0.001 0.001 a Average assuming the same number of respondents in each country. b Using ANOVA (geometric mean) and chi-squared tests (categorical variable <16,16-20,>20) for mean age of first smoking, and chi-squared test for no. of cigarettes smoked (<=20,>20), time to first cigarette (first hour,later) and dependency (light dependence, moderate/heavy dependence) c Using t-tests d Using chi-squared tests

170

Age at initiation The majority of male smokers reported that they began smoking before the age of 20 years, and, on average, a quarter reported that they started in childhood (Table 4-3). Far fewer women started in childhood and a sizeable portion started over the age of 20; for example, 86% of women in Armenia and more than 40% of those in Georgia, Kyrgyzstan and Moldova reported that they started smoking after this age. These gender differences were significant in all countries.

Differences were also observed between countries; in Belarus, Kazakhstan, Russia and Ukraine the geometric mean age of first smoking was under 18 in men and under 20 in women, compared with older ages elsewhere. Overall between country differences were significant in both genders (p<0.001) yet Bonferroni multiple comparisons show that significant differences in women existed only when comparing Armenia with countries other than Georgia and Moldova (p<0.01, data not shown). Among men, significantly younger ages of initiation were seen in Russia and Ukraine compared with Armenia, Georgia, Kyrgyzstan and Moldova, in Belarus compared with Armenia and Kyrgyzstan, and in Kazakhstan compared with Kyrgyzstan (all p<0.01, data not shown).

Amount smoked and nicotine dependence Men tended to smoke more cigarettes than women, with the majority smoking 10 or more cigarettes per day while most women smoked fewer than 10. Between-gender differences in the proportion of respondents smoking more than 20 cigarettes per day reached significance only in Belarus, Kazakhstan, Russia and Ukraine (p<0.001).

The majority of smokers smoked their first cigarette within an hour of waking although in all countries bar Georgia a far higher proportion of men than women did so (p<0.01). Men are therefore more likely to be moderately to severely nicotine dependent although gender differences were significant only in Belarus, Kazakhstan, Russia and Ukraine.

Determinants of smoking For the sake of brevity, only the multivariate analyses will be presented in full here (Table 4-4 & Table 4-5) although the age-adjusted results, where different, will be described. Age and religion were significant predictors of smoking in both genders. A significant decline in smoking with increasing age was seen in men from all countries and in females from all countries except Armenia. Kazakh and Kyrgyz

171

Muslim respondents had a much lower risk of smoking compared with other respondentsq. Marital status was related to smoking in both genders but in different ways. In males, being single was associated with a lower rate of smoking in Ukraine and Kazakhstan. In women, those who were separated or divorced tended to have a higher risk of smoking. This relationship was seen in the multivariate model when all countries were examined together and in the age-adjusted model in Kazakhstan, Kyrgyzstan and Ukraine.

Thereafter, the variables associated with smoking differed between the genders. In men, educational achievement was significantly inversely related to smoking in six countries, Belarus, Georgia, Kazakhstan, Moldova, Russia and Ukraine (p-value for trend <0.01). Men with poorer economic status were less likely to smoke in Russia although in the age adjusted analysis this association was also seen in Kyrgyzstan and Ukraine. A lack of social support was significantly associated with smoking in Kyrgyzstan and Russia, but the association seen in the age adjusted analysis was reduced to borderline significance in Kazakhstan.

In women no significant association was seen with education, economic position or social support, even when all countries were combined. The most notable association in women was the higher risk of smoking in women living in urban areas. This was seen in all countries although in the age-adjusted analysis the association had not been significant in Kyrgyzstan.

When data from all countries were pooled, being older, single, Muslim, more highly educated, in a better economic situation, and with more social support was associated with a lower likelihood of smoking in males. In women, those who were older, living in less urbanised areas, and Muslim had a lower risk of smoking, while women who were divorced, separated or widowed were more likely to smoke than married women.

Finally, we can also see that once differences in demographic and socio-economic variables were taken into account, between-country differences in smoking prevalence remained (Figure 4-4 & Figure 4-5) albeit with some minor differences from the unadjusted prevalence data presented above. Men from Moldova, Ukraine, Belarus and Georgia were less likely to smoke than Russian men while those from q The number of Muslim respondents was too small to assess this relationship in other individual countries.

172

Kazakhstan were 32% more likely to smoke. In females, compared with respondents from Russia, those from Armenia, Moldova, Georgia and Kyrgyzstan were far less likely to smoke, while those from Kazakhstan and Belarus had a somewhat lower rate of smoking. In addition, women in Armenia, Moldova and Georgia were also less likely to smoke than those in Ukraine and Belarus.

173

Table 4-4 Determinants of smoking in males: odds ratios for the likelihood of smoking adjusted for all other factors. AM BY GE KZ KG MD RU UA All countries OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI n=793 n=836 n=893 n=850 n=878 n=876 n=1803 n=885 n=7660 Country p<0.0001 Russia ------1.00 Armenia ------AM 0.97 0.76 ; 1.24 Belarus ------BL 0.75 0.59 ; 0.94 Georgia ------GE 0.78 0.62 ; 0.99 Kazakhstan ------KZ 1.32 1.02 ; 1.71 Kyrgyzstan ------KG 0.76 0.57 ; 1.03 Moldova ------MO 0.42 0.33 ; 0.53 Ukraine ------UK 0.72 0.57 ; 0.90

Age group p<0.0001 p<0.0001 p<0.0001 p=0.0001 p<0.0001 p<0.0001 p<0.0001 p<0.0001 p<0.0001 18-29 years 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 30-39 years 1.30 0.59 ; 2.90 1.27 0.64 ; 2.55 1.10 0.55 ; 2.20 1.18 0.60 ; 2.29 1.04 0.57 ; 1.87 0.57 0.29 ; 1.15 1.05 0.63 ; 1.77 0.57 0.24 ; 1.33 1.00 0.79 ; 1.26 40-49 years 0.77 0.34 ; 1.75 0.95 0.48 ; 1.88 0.83 0.40 ; 1.71 0.77 0.38 ; 1.55 0.65 0.35 ; 1.18 0.38 0.19 ; 0.76 0.83 0.49 ; 1.40 0.32 0.14 ; 0.75 0.69 0.55 ; 0.88 50-59 years 0.68 0.26 ; 1.81 1.06 0.50 ; 2.28 0.54 0.25 ; 1.17 0.65 0.30 ; 1.43 0.49 0.23 ; 1.07 0.29 0.14 ; 0.61 0.58 0.34 ; 1.00 0.34 0.14 ; 0.85 0.57 0.44 ; 0.74 60+ years 0.25 0.11 ; 0.57 0.34 0.17 ; 0.71 0.22 0.10 ; 0.47 0.34 0.16 ; 0.74 0.15 0.07 ; 0.34 0.13 0.06 ; 0.28 0.23 0.14 ; 0.38 0.10 0.04 ; 0.24 0.22 0.17 ; 0.28

Area p=0.39 p=0.24 p=0.15 p=0.25 p=0.93 p=0.52 p=0.56 p=0.69 p=0.61 Capital/region. capital 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Other cities 0.79 0.44 ; 1.42 1.31 0.83 ; 2.07 1.15 0.62 ; 2.14 1.13 0.66 ; 1.93 0.91 0.46 ; 1.82 0.76 0.40 ; 1.46 0.97 0.70 ; 1.34 1.15 0.73 ; 1.81 1.06 0.89 ; 1.25 Village 0.80 0.50 ; 1.27 1.28 0.80 ; 2.04 0.79 0.51 ; 1.22 1.35 0.85 ; 2.16 0.99 0.60 ; 1.62 0.87 0.56 ; 1.36 0.86 0.60 ; 1.24 1.14 0.70 ; 1.84 1.00 0.85 ; 1.16

Marital status p=0.02 p=0.39 p=0.93 p=0.007 p=0.41 p=0.38 p=0.06 p=0.003 p<0.0001 Married/common law 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Single 0.45 0.21 ; 0.96 0.83 0.43 ; 1.60 0.91 0.48 ; 1.73 0.47 0.24 ; 0.92 0.79 0.43 ; 1.45 0.67 0.32 ; 1.41 0.64 0.39 ; 1.03 0.39 0.17 ; 0.86 0.65 0.52 ; 0.81 Separated/divorced 1.18 0.53 ; 2.65 1.29 0.70 ; 2.36 1.02 0.45 ; 2.30 0.63 0.29 ; 1.37 0.72 0.30 ; 1.76 0.97 0.52 ; 1.79 1.03 0.65 ; 1.65 1.36 0.77 ; 2.41 1.06 0.85 ; 1.33 /continued

174

Table continued AM BY GE KZ KG MD RU UA All countries OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI n=793 n=836 n=893 n=850 n=878 n=876 n=1803 n=885 n=7660 Religion p=0.0028 p=0.0004 p<0.0001 Non muslim - - - 1.00 1.00 - - - 1.00 Muslim ---0.57 0.37 ; 0.85 0.50 0.28 ; 0.91 -- -0.65 0.51 ; 0.84

Education level p=0.08 p<0.0001 p=0.002 p<0.0001 p=0.21 p=0.003 p<0.0001 p<0.0001 p<0.0001 Secondary or less 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Sec. vocational / some 0.72 0.45 ; 1.17 0.73 0.47 ; 1.13 0.56 0.36 ; 0.89 0.74 0.47 ; 1.17 0.85 0.56 ; 1.30 0.57 0.36 ; 0.88 0.70 0.51 ; 0.98 0.54 0.35 ; 0.85 0.69 0.60 ; 0.80 higher educ. Completed higher 0.66 0.38 ; 1.15 0.37 0.22 ; 0.65 0.60 0.37 ; 0.96 0.30 0.17 ; 0.53 0.71 0.43 ; 1.18 0.64 0.36 ; 1.15 0.31 0.21 ; 0.45 0.41 0.25 ; 0.68 0.47 0.39 ; 0.56 education

Economic situation of the p=0.47 p=0.06 p=0.15 p=0.48 p=0.02 p=0.08 p=0.007 p=0.013 p<0.0001 family this year Bad/very bad 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Average 0.99 0.64 ; 1.51 0.65 0.41 ; 1.04 0.76 0.52 ; 1.12 1.26 0.75 ; 2.13 0.60 0.35 ; 1.02 0.71 0.47 ; 1.05 0.73 0.54 ; 0.996 0.67 0.45 ; 0.99 0.77 0.66 ; 0.88 Good/very good 0.59 0.20 ; 1.78 0.69 0.36 ; 1.30 1.23 0.35 ; 4.27 1.29 0.66 ; 2.52 0.52 0.28 ; 0.995 0.83 0.41 ; 1.69 0.59 0.35 ; 0.97 0.49 0.20 ; 1.21 0.70 0.56 ; 0.89

Lack of social support p=0.94 p=0.65 p=0.21 p=0.04 p=0.007 p=0.81 p=0.009 p=0.22 p=0.002 No lack 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Some lack 0.94 0.58 ; 1.52 1.20 0.72 ; 1.99 0.62 0.30 ; 1.28 1.53 0.93 ; 2.52 1.21 0.77 ; 1.91 1.00 0.62 ; 1.61 1.53 1.06 ; 2.21 1.40 0.85 ; 2.29 1.22 1.03 ; 1.44 Severe lack 0.97 0.52 ; 1.81 1.05 0.59 ; 1.88 1.10 0.53 ; 2.26 1.54 0.79 ; 2.99 2.55 1.17 ; 5.56 0.86 0.47 ; 1.58 1.22 0.79 ; 1.88 1.06 0.57 ; 1.98 1.20 0.98 ; 1.48

175

Table 4-5 Determinants of smoking in females: odds ratios for the likelihood of smoking adjusted for all other factors. AM BY GE KZ KG MD RU UK All countries OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI n=1144 n=1071 n=1041 n=1070 n=1076 n=1060 n=2222 n=1399 n=10066 Country p<0.0001 Russia ------1.00 Armenia ------AM 0.10 0.06 ; 0.17 Belarus ------BL 0.72 0.53 ; 0.97 Georgia ------GE 0.35 0.24 ; 0.50 Kazakhstan ------KZ 0.66 0.47 ; 0.93 Kyrgyzstan ------KG 0.49 0.30 ; 0.79 Moldova ------MO 0.24 0.15 ; 0.38 Ukraine ------UK 0.77 0.58 ; 1.02

Age group p=0.04 p<0.0001 p=0.004 p<0.0001 p=0.0012 p=0.003 p<0.0001 p<0.0001 p<0.0001 18-29 years 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 30-39 years 5.14 0.52 ; 50.48 0.79 0.37 ; 1.71 2.04 0.68 ; 6.06 0.52 0.24 ; 1.16 1.81 0.57 ; 5.73 1.24 0.37 ; 4.17 0.70 0.45 ; 1.10 0.31 0.16 ; 0.59 0.72 0.55 ; 0.93 40-49 years 6.97 0.74 ; 66.04 0.45 0.20 ; 1.02 1.46 0.46 ; 4.65 0.53 0.23 ; 1.20 0.94 0.27 ; 3.31 0.33 0.08 ; 1.46 0.32 0.19 ; 0.53 0.15 0.07 ; 0.34 0.41 0.31 ; 0.55 50-59 years 3.77 0.29 ; 49.02 0.09 0.03 ; 0.35 0.60 0.14 ; 2.54 0.09 0.02 ; 0.48 0.50 0.10 ; 2.47 0.53 0.11 ; 2.56 0.29 0.17 ; 0.50 0.14 0.06 ; 0.31 0.25 0.18 ; 0.36 60+ years 1.20 0.08 ; 17.64 0.02 0.00 ; 0.13 0.49 0.12 ; 1.97 0.01 0.00 ; 0.13 0.09 0.01 ; 0.60 0.10 0.01 ; 0.70 0.04 0.02 ; 0.09 0.01 0.00 ; 0.05 0.05 0.03 ; 0.07

Area p=0.004 p<0.0001 p=0.0012 p<0.0003 p=0.015 p<0.0001 p<0.0001 p<0.0001 p<0.0001 Capital/region. capital 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Other cities 0.10 0.01 ; 1.50 0.74 0.42 ; 1.32 0.36 0.12 ; 1.07 1.02 0.50 ; 2.07 0.33 0.08 ; 1.30 0.73 0.24 ; 2.21 0.68 0.47 ; 0.98 0.48 0.27 ; 0.84 0.60 0.48 ; 0.76 Village 0.13 0.02 ; 0.94 0.22 0.09 ; 0.55 0.32 0.12 ; 0.82 0.32 0.15 ; 0.71 0.36 0.12 ; 1.04 0.08 0.02 ; 0.26 0.41 0.26 ; 0.65 0.30 0.16 ; 0.57 0.31 0.23 ; 0.40

Marital status p=0.02 p=0.06 p=0.19 p=0.014 p=0.13 p=0.61 p=0.08 p=0.04 p<0.0001 Married/common law 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Single 2.68 0.49 ; 14.63 1.97 0.94 ; 4.16 0.75 0.25 ; 2.29 0.75 0.30 ; 1.89 0.67 0.16 ; 2.83 0.71 0.14 ; 3.65 0.98 0.60 ; 1.58 1.04 0.52 ; 2.11 1.04 0.78 ; 1.38 Separated/divorced 3.30 1.00 ; 10.86 1.30 0.63 ; 2.66 1.73 0.70 ; 4.24 2.11 1.01 ; 4.41 1.92 0.71 ; 5.15 1.38 0.46 ; 4.15 1.40 0.94 ; 2.08 1.87 0.98 ; 3.58 1.54 1.21 ; 1.97 /continued

176

Table continued AM BY GE KZ KG MD RU UA All countries OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI OR 99% CI n=793 n=836 n=893 n=850 n=878 n=876 n=1803 n=885 n=7660 Religion p=0.003 p<0.0001 p<0.0001 Non muslim - - - 1.00 1.00 - - - 1.00 Muslim ---0.44 0.21 ; 0.90 0.09 0.03 ; 0.25 ---0.26 0.16 ; 0.43

Education level p=0.81 p=056 p=0.35 p=0.24 p=0.96 p=0.10 p=0.02 p=0.09 p=0.19 Secondary or less 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Sec. vocational / some 1.15 0.31 ; 4.28 0.82 0.45 ; 1.50 0.82 0.32 ; 2.11 0.64 0.32 ; 1.31 0.92 0.34 ; 2.48 0.45 0.16 ; 1.25 0.88 0.60 ; 1.28 1.26 0.70 ; 2.27 0.87 0.69 ; 1.09 higher educ. Completed higher 1.40 0.36 ; 5.45 0.75 0.34 ; 1.65 1.28 0.54 ; 3.00 0.66 0.29 ; 1.53 0.88 0.29 ; 2.69 0.48 0.14 ; 1.61 0.60 0.38 ; 0.96 1.80 0.89 ; 3.62 0.86 0.66 ; 1.12 education

Economic situation of the p=0.89 p=0.24 p=0.94 p=0.10 p=0.26 p=0.71 p=0.05 p=0.13 p=0.02 family this year Bad/very bad 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Average 1.10 0.36 ; 3.33 0.69 0.37 ; 1.29 1.10 0.54 ; 2.23 0.55 0.27 ; 1.13 1.90 0.69 ; 5.27 0.99 0.39 ; 2.52 0.71 0.50 ; 1.02 0.67 0.40 ; 1.15 0.80 0.64 ; 0.98 Good/very good 1.65 0.10 ; 28.13 0.99 0.38 ; 2.59 0.98 0.13 ; 7.22 0.68 0.26 ; 1.80 1.88 0.45 ; 7.92 0.53 0.07 ; 4.10 0.85 0.47 ; 1.53 0.61 0.21 ; 1.81 0.85 0.59 ; 1.22

Lack of social support p=0.65 p=0.04 p=0.13 p=0.36 p=0.42 p=0.63 p=0.69 p=0.66 p=0.16 No lack 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 Some lack 0.62 0.14 ; 2.67 1.84 0.95 ; 3.57 0.78 0.24 ; 2.55 0.97 0.45 ; 2.13 1.54 0.58 ; 4.10 1.45 0.54 ; 3.89 1.07 0.71 ; 1.61 0.78 0.40 ; 1.56 1.13 0.88 ; 1.44 1.15 0.25 ; 5.28 1.65 0.64 ; 4.22 0.23 0.03 ; 1.55 1.57 0.67 ; 3.69 1.57 0.45 ; 5.50 1.12 0.28 ; 4.52 1.19 0.68 ; 2.08 0.96 0.40 ; 2.30 1.24 0.89 ; 1.72 Severe lack

177

Figure 4-4 Odds (and 99% confidence interval) of smoking in men having adjusted for age, area of residence, marital status, religion, education, economic situation and level of social support

1.80

1.60

1.40

1.20

1.00

0.80

0.60

0.40

0.20

0.00 MO UA BL KG GE AM RU KZ

178

Figure 4-5 Odds (and 99% confidence interval) of smoking in women having adjusted for age, area of residence, marital status, religion, education, economic situation and level of social support

1.20

1.00

0.80

0.60

0.40

0.20

0.00 AM MO GE KG KZ BL UA RU

DISCUSSION Strengths of these surveys These surveys of over 18,000 individuals provide important new data on the prevalence and determinants of smoking in eight countries representing more than four-fifths of the population of the former Soviet Union. For some countries they provide the first accurate, country-wide data on smoking prevalence. The LLH data have a number of advantages over existing data. First, having been collected using standardised methods, they provide some of the first truly comparative data for countries of the FSU other than the Baltic states.115,116 Second, having been based on samples of the general adult population of each participating country, of which they were broadly representative, the findings are generalisable. Third, by having clearly defined sample characteristics they offer advantages over data available in public databases.

179

Study limitations Response rates were relatively high and the samples broadly representative of their study populations. However, although male under-representation in Armenia and Ukraine will not affect gender specific rates, the urban/rural differences may overestimate prevalence rates in Kyrgyzstan, where urban areas were over-represented, while the opposite was true in Armenia. This is however only likely to affect data on females to any extent128,129,130 The age group disparities noted were minor but will tend to underestimate smoking prevalence. In addition the surveys were based on self-reported smoking status with no independent biochemical validation and may thus have been affected by reporting bias. Although there is some concern that self reported smoking status may under-estimate smoking and the amount smoked, studies in the west suggest it is a sensitive and specific measure and that interviewer-administered questionnaires provide more accurate responses than self-completed questionnaires. 336 The only study in the FSU that addresses this issue found that among those claiming to be non-smokers, 13% (48/368) of women and 17% (12/375) of men in rural north-west Russia were, according to blood cotinine levels, likely to be smokers compared with only 2% of each gender in Finland.337 Given the far lower prevalence of smoking among women this had a disproportionately large impact on the reported female smoking prevalence. Although our questionnaires were administered by interviewers in the respondents’ homes, potentially making it harder for respondents to deny smoking, our study may underestimate smoking prevalence particularly in women in areas where smoking remains culturally unacceptable. Due to limitations on questionnaire size, I was unable to include questions on smokeless tobacco use, which is fairly common place in parts of the FSU, mainly Azerbaijan, Tajikistan and Turkmenistan. Although chewing tobacco is used in some southern parts of Kyrgyzstan, cigarettes are the main form of tobacco used here and in all the other countries surveyed.330,338

Although the overall sample size in each country was relative largely, the small size of certain population subgroups (particularly in women where the number of smokers was smaller) reduced the power to detect significant differences. Furthermore, data were

180

collected only on individuals aged 18 years and over which precludes the exploration of smoking correlates in adolescents, among the main targets of the TTCs in this region.

Findings Smoking prevalence The study confirms that male smoking rates in this region are among the highest in the world and higher than the maximum rates recorded in the US at the peak of its epidemic, with rates over 50% seen in all countries except Moldova and reaching 60% or more in Armenia, Kazakhstan and Russia. Elsewhere in Europe at the time of this survey rates over 50% were only seen in Turkey (51%) and Slovakia (56%) and worldwide less than 20 countries report rates over 60%.118

In men the lower prevalence of current smokers and higher prevalence of never and ex- smokers in those over 60 is likely to reflect the fact that some men in this age group have given up, plus the disproportionate number of premature deaths among current smokers compared with never and ex-smokers. However, there is also known to be a cohort effect in the FSU with those who were teenagers between 1945 and 1953 carrying forward lower smoking rates as cigarettes, like other consumer goods, were in short supply in the period of post-war austerity under Stalin.180,195

Compared with male smoking patterns, smoking in women is far less common, varies more between countries and has a different age-specific pattern. Although ever smoking rates are under 4% in the over 60s in all eight countries, in the four countries with the highest female smoking rates (Belarus, Kazakhstan, Russia and Ukraine), smoking is now significantly more common in younger generations with risk ratios between the youngest and oldest age-groups of 12.2 to 37.3 compared with 1 to 5.5 in the other four countries.

Comparison with previous prevalence data is problematic as much of what exists is fragmentary, of uncertain quality and often not nationally representative with urban samples likely to overestimate smoking in women (see Chapter 1, Section 3.3). These problems apply particularly to central Asian and Caucasian states although limited data from Armenia and Moldova dated between 1998 and 2001 suggest little change in smoking prevalence329,118 while those from Kazakhstan suggest a small increase from

181

the 60% male and 7% female prevalence rates recorded in 1996.329,r More data are available for Belarus, Russia and Ukraine. These suggest that smoking rates in men have changed little in recent years329,129,130,339 although in Russia, they appear to have risen from approximately 40-50% in the 1970s and 1980s329,67,68 to around 60% in the mid 1990s, with little subsequent change. In women, rates appear to have increased in all three countries.329,130 In Belarus for example rates in women have climbed steadily from under 5% in the mid 1990s to a maximum of 12% in this survey.130,329 Data for Russia as whole suggest that prevalence in women has risen from around 10% in the early 1990s to 15% now. Pre-transition data on women are confined to Moscow or other areas and whilst not directly comparable suggest that rates have been rising since the 1970s but most notably through the 1990s.67,68,128,340 Similarly in Ukraine historical data from Kiev show a steady rise in smoking among women from the mid 1970s to 1990s while male smoking rates barely changed, hovering around 50%. More recent national data suggest male smoking then rose slightly to reach approximately 57% by the turn of the century, 129,339 suggesting that the figure of 52.5% in the present survey could represent a downturn, although further data will be needed to confirm this. In women, the only nationally comparative data is my previous survey which found a rate of 10% in 2000 comapred to 11.1% here. Although other surveys reported rates of 14% in 2000 and 2001, the difference is likely to be accounted for by their slightly younger age samples.339

Smoking determinants The relationship between age and smoking seen in both genders in this study is consistent with previous surveys conducted in Belarus, Russia, Ukraine and the Baltic Republics,128,129,130,131,341 and is observed in most of western Europe.67,342,343 In males, more socially disadvantaged respondents were particularly likely to smoke, i.e. those with a lower education level, in poorer economic situation, and with less social support; a tendency also reported in previous surveys from Belarus, Russia and Ukraine although the markers of disadvantage varied among studies.128,129,130,339,341 More generally, as noted in Chapter 1, it is these men whose health appears to have suffered most since the

r The only recent data for Kyrgyzstan are from a 1997 casual sample of clinic attendees in Bishkek clinic attendees in the capital Bishkek (personal correspondence Chinara Bekbasarova). In Georgia, previous reports come from small surveys in Tbilisi which cannot be directly compared with our results.124,125

182

collapse of the former Soviet Union.58,344,345,57 The inverse relationship between smoking and educational achievement in males was particularly striking and is consistent with results from other surveys conducted in the Baltic states and in Belarus.130,131 Previously published results from Russia and the Ukraine are less consistent; while one study also showed a significant inverse trend with education,341 another showed an inverse trend that did not reach significance,337 some showed no clear association128,129,346 and others reported a lower prevalence in those with primary compared with higher education (although significance was not assessed and having only primary education in the USSR was often a consequence of other health problems that limited participation in society).339 It is well known that the association of education with material rewards and status was less consistent in the former Soviet Union than in western countries347 and, seemingly as a consequence, the relationship between education and health seen in the west is not routinely observed in the former Soviet Union.60

In this study, we observed that men reporting less social support were more likely to smoke than other men. These results tend to agree with those of studies indicating that men with more social ties have a lower risk of death compared with other men.348,349 Similarly our previous work in Ukraine suggested that more socially isolated men were more likely to smoke129 and that good family relations offered protection against poor self-perceived health.60

In women, the relationship between disadvantage and smoking was not as clear. This could be explained by the fact that the epidemic is at an earlier point of development in women in the region and smoking tends to be taken up first by those in more advantageous positions. Unlike with men, it was women living in more urbanised areas and those who were separated, divorced or widowed who reported higher rates of smoking. The inverse relationship between urbanisation and smoking prevalence in women is consistent with previous observations from Belarus, Latvia, Lithuania and Russia,128,129,131 and suggests lower prevalence rates in women living in more traditional settings, with higher rates in those with more exposure to western influences and to recent aggressive advertising campaigns by western tobacco companies.86 Internal tobacco industry documents reviewed in Chapters 6, 8 and 11 show that female smokers were seen as an important potential market as they were more likely to smoke

183

international filter brands, that brands were marketed directly to women, and that marketing was targeted, in the first instance at least, at large cities and to young opinion leaders. Moreover, as Chapters 8 and 11 illustrate, the documents predict that such marketing efforts will lead to increased smoking rates, particularly amongst women.

The tobacco epidemic As outlined in Chapter 1, Section 4.4, a four stage model of the smoking epidemic has been outlined, based on observations in the west.172 It describes the initial rise in male smoking, followed by the rise in female smoking one to two decades later. Each then plateaus and falls with tobacco related mortality rising to a peak decades later. Our findings suggest that the tobacco epidemic may have developed differently in the FSU. Male smoking has a long history in this region. The first accounts of tobacco smoking in Russia date from the 17th century,61 papirossi were first mentioned in184461 and cigarette factories first constructed later in the nineteenth century.61,63 Historical data on smoking67 and high male tobacco-related mortality rates145 suggest that male smoking has been high for some time, failing to decline after a peak as the model would predict. Female smoking remains relatively uncommon, and rates have been far slower to rise than would be expected given male rates and trends observed in the west. Indeed, it appears that it is only since the mid to late 1990s, when the TTCs entered with their carefully targeted marketing strategies,86,84,14 that women really started smoking in substantial numbers. Therefore, although the exact stage of the epidemic varies slightly between the countries of the FSU, overall we suggest that men have remained between stages 3 and 4 of the epidemic, with high rates of both smoking and mortality, while women in some countries are in stage 1 and in others in stage 2, the latter with more rapidly rising smoking rates. Female lung cancer rates have yet to increase and although rates of cardiovascular disease have been increasing, the reasons for this are complex and numerous.

Potential explanations for between country differences Between gender and inter-country differences in smoking prevalence described above are also reflected in other smoking habits: men are more likely than women to start smoking when young, smoke more heavily and be nicotine dependent. Overall, two groupings of countries appear to emerge from the between country comparisons -

184

Belarus, Kazakhstan, Russia and Ukraine on one hand and Armenia, Georgia, Kyrgyzstan and Moldova on the other. In addition to exhibiting higher female smoking rates and more pronounced age specific trends, the former group tend to have lower ages of smoking uptake (particularly when compared with Armenia, Georgia and Moldova) and more marked gender differences in the number of cigarettes smoked per day and levels of nicotine dependency.

The between country differences observed in this study suggest that smoking patterns in Armenia, Georgia, Moldova and Kyrgyzstan are more traditional than those in Belarus, Kazakhstan, Russia and Ukraine. This could be explained by the differing degree of TTC penetration in these countries (see Chapter 2).350 The Moldovan industry remains a state owned monopoly and although the Georgian and Armenian industries have been privatised, this change was rather recent (post 1997) and none of the major TTCs invested directly. Kazakhstan, Russia and Ukraine, by contrast, saw major investments from most of the major tobacco companies in the early 1990s onwards. Belarus, which retains a state owned monopoly, and Kyrgyzstan, where the German manufacturer Reemtsma invested, would therefore appear to be exceptions, with Belarus more typical of the countries with TTC investments and Kyrgyzstan of the countries without such investments. In Belarus, however, the state manufacturer has only a 40% market share, with an additional 40% made up of smuggled and counterfeit brands. The importance the TTCs attach to this illegal market is illustrated in the fact that, despite having little official market share,350 British American Tobacco (BAT) and Philip Morris have the highest outdoor advertising expenditure and the ninth and tenth highest television advertising expenditures of all companies operating in Belarus.271 As in Ukraine and Russia, tobacco is the product most heavily advertised outdoors and the fourth most advertised product on television (there are now restrictions on television advertising in Ukraine and Russia).210,271 It is clear therefore that with the continuing (if so far fruitless) discussion of possible reunification with Russia, the TTCs treat Belarus as an important extension of the Russian market.351

Kyrgyzstan differs from the other countries in which there have been TTC investments in that these investments occurred later (1998) and gave Reemtsma a manufacturing monopoly.350 However, Kyrgyzstan also differs from Belarus, Kazakhstan, Ukraine and

185

Russia in its lower level of development and industrialisation and its larger rural and Muslim populations. Most importantly, this study shows that between country differences, for example in levels of education, socio-economic status and religion, do not account entirely for the differences in smoking prevalence that are identified; the multivariable analysis shows that when urban status, socio-economic factors and religion are accounted for, the differences in female smoking rates between Kyrgyzstan and Kazakhstan, Belarus and Ukraine disappear although those with Russia remain. Moreover, having controlled for these between country differences, the differences in female smoking in Armenia and Moldova remain significantly different from all four countries in the other group, whilst smoking levels in Georgia differ significantly from all bar Kazakhstan (Figure 4-5).

Thus, our findings suggest that levels of smoking prevalence, particularly amongst women, are associated with the extent of market entry by the transnational tobacco industry. This is consistent with evidence from other parts of the world showing how the entry of the industry into formerly closed markets is associated with an increase in smoking.104,352

CONCLUSIONS The survey findings, combined with data on disease burden,180,145 confirm that the long- standing high smoking rates in men continue unabated. Amongst women, smoking in Armenia, Georgia, Kyrgyzstan and Moldova remains relatively uncommon and does not appear to have increased significantly as judged by rates in younger compared with older generations or by comparison with previous data. By contrast, female smoking rates in Belarus, Ukraine, Kazakhstan and Russia are higher, show an increase from previous surveys and the age specific rates suggest an ongoing rise in younger generations. It is unlikely to be a coincidence that the higher rates are observed in countries with the most active TTC presence.

186

CHAPTER 5 Tobacco document provenance and methodology

BACKGROUND The provenance of the tobacco industry documents Litigation against tobacco companies commenced in the mid 1950s but was, until recently, totally unsuccessful. Fortunes changed however once internal industry documents provided by whistleblowers started to reveal the extent of corporate misbehaviour353 and when litigants switched their actions from individual tort cases to class action tort suits, state health care reimbursement cases, and filings by health insurers.354

Mississippi, Minnesota, West Virginia and Florida were the first states to file cases, in 1994 and 1995. They were soon followed by many others and by June 1997 forty states were bringing suits and Blue Cross and other health insurers were launching parallel actions. In 1998 a number of major settlements were reached. The industry settled individually with Mississippi, Florida, Texas and Minnesota, the states seen to present the greatest threat, and in November with the 46 remaining states in the Master Settlement Agreement.355

Two of these settlements led to the public release of internal industry documents from six tobacco companies (American Tobacco Company, British American Tobacco Company, Brown & Williamson Tobacco Corporation, Lorillard Tobacco Company, Philip Morris Incorporated and RJ Reynolds Tobacco Company) and two industry organisations (the Council for Tobacco Research and the Tobacco Institute).356 The Minnesota Consent Judgement ordered the companies involved in the Minnesota trial to provide public access to the millions of pages of their documents produced as part of the legal discovery process and held at two depository sites established during the trial, one in Minnesota, US (holding the American based tobacco company documents) and the other in Guildford, England (housing BAT’s documents).357 These depositories are open for ten years. The Minnesota settlement also required continued public disclosure of tobacco company documents produced in subsequent smoking and health litigation in the US for ten years.

187

The Master Settlement Agreement (MSA) required the companies, other than BAT or the Liggett Group,358 to post their documents on public websites and maintain them there until June 30, 2010.359 In addition to these industry run websites, other document websites have subsequently sprung up to provide access to the US company documents in a more user friendly format that enables searching of all company documents at one time and will provide access in the long-term. The most notable sites are Tobacco Documents Online (http://tobaccodocuments.org/) and Legacy Tobacco Documents Library (http://legacy.library.ucsf.edu/).

Access to the tobacco industry documents: Guildford vs Minnesota The stipulation of internet access for the US tobacco company documents but not for BAT’s documents is only one of many differences in access to these important resources. In contrast to its conditions imposed on the U.S. tobacco companies, the Minnesota settlement was vague on the precise terms of public access for the BAT documents housed at Guildford. Thus, whilst the Minnesota depository is run independently of the tobacco industry, the Guildford depository is directly administered by BAT. Its management and operation has been problematic from the start with BAT delaying its opening until February 1999, almost a year after the Minnesota Depository opened, and doing so only after persistent lobbying.

Subsequently public access to the documents has been hindered by BAT in a number of ways that have made research difficult. BAT limited access to a single organisation (comprising no more than six individuals) at any one time, opened the depository only six hours per day and required visitors to book through its lawyers in advance in one week timeslots. As a result, as one of the earliest researchers to visit the depository in 2000, I was waiting for up to five months to book a one week slot.360 The Minnesota Depository, conversely, allows multiple groups to visit at any one time and is open twelve hours per day.

Once in the depository, conditions are hostile, with researchers watched through a one- way mirror and surveillance cameras recording movements. Searching its contents also proved problematic. Unlike the Minnesota Depository, the Guildford Depository documents are indexed and searchable only by file rather than by document, thereby

188

making it impossible to search the Depository's database for individual documents within the collection. Again, unlike in Minnesota, copy machines are not available and visitors must order photocopies of documents, delivery of which, through the course of my work, took anywhere from a few months to up to a year.

Recent BAT documents produced through subsequent litigation and now housed in Minnesota confirm that the company and its lawyers acted to hinder access to and research on the documents.361,362 They show that BAT undertook detailed surveillance of researchers including the monitoring of electronic database searches. They outline how BAT lawyers ranked the sensitivity of files requested and altered documents that were potentially embarrassing. For example a document from BAT’s lawyer’s, Lovells, dated 2000 refers to the contents of a document requested by a team of researchers (including myself) noting its sensitive nature “due to references to marketing to illiterate, low income 16-year olds (reference to 16-year olds changed in manuscript to 18-year olds).”363 Other documents obtained from Guildford and those submitted in evidence in various legal cases suggest that BAT or lawyers acting on its behalf have deliberately destroyed sensitive and potentially damaging documents.364,365,366,367 In this regard, our close monitoring of file numbers in the depository justifies concern: the total number of files held in May 2004 was 181 (approximately 36,000 pages) fewer than the 40,784 files reported to the Health Select Committee by BAT in January 2000.368

BAT does not have to make public documents that contain legally privileged, trade secret or personal material. It maintains a separate database at the depository of documents on which it has claimed legal privilege. There is however no way of checking the veracity of such claims. The fact that BAT’s lawyers have on some occasions given me copies of a document that they later claim as privileged suggest that this term is at best inconsistently applied. It has also been suggested that BAT’s claims of privilege have been based upon inference rather than evidence of authorship.361

The Health Select Committee, recognising the problems of access and archive integrity, recommended that BAT make the contents of the depository available via the internet, a recommendation endorsed by the British Government. BAT, despite evidence that it has substantial imaging capability at Guildford, has refused to comply, denying it has such

189

capacity.361 For these reasons other efforts have been made to improve access to the Guildford documents.

Improving access Since the start of this PhD, small numbers of BAT documents acquired by tobacco control activists and researchers visiting Guildford have been posted on the World Wide Web, mainly at the University of California San Francisco (http://library.ucsf.edu/tobacco/batco/) and Tobacco Documents Online websites (http://tobaccodocuments.org/).

In addition, a group of us at LSHTM, in conjunction with collaborators in the US, launched the Guilford Archiving Project (GAP)369 to ensure long-term access to the important collection of BAT documents. The GAP project entails copying, scanning and indexing all 8 million pages of BAT documents held at Guildford and then placing them on a publicly accessible website in perpetuity. The project was publicly launched in May 2004362 and the first documents went online in October 2004.370

Use of tobacco industry documents The estimated 40 million pages of internal industry records have revolutionised tobacco control by providing unique insights into the thinking, strategies and tactics of the TTCs.371 For example the extent to which the industry opposes or espouses certain tobacco control measures provides a useful test of that measure’s likely effectiveness. Experts argue that document analysis has had a substantial impact on the tobacco policy- making process, both domestically and internationally372 and a WHO report recently described the documents as providing “the most important tool for reframing the debate in a member state”.373

Analysis of these tobacco industry documents has led to a steady growth in papers examing industry tactics and strategies related to both research and policy. 374 Most papers have focused on the US and by 2003 there had been few analyses of TTC activities in international markets other than China,375 Latin America376 , and a few countries in Europe,377,378,379 as well as a review of tobacco industry influence on the WHO.380 The absence of any analysis of TTC conduct in low income countries or in the FSU is notable, particularly given the TTC’s focus in recent years.

190

In addition, due to their far greater accessibility, analyses have focused on the documents from American companies, largely those of Philip Morris, RJ Reynolds and their related organisations, whilst documents belonging to BAT, the world’s second largest tobacco company381 have been relatively neglected.382

METHODOLOGY Documents used Given the proximity of the London School of Hygiene and Tropical Medicine to Guildford, Surrey, the absence of thorough analyses of this depository, the important role that BAT plays in the global market (it boasts the most extensive global presence) and the fact that BAT is a British company, the documentary analyses in this thesis focus on BAT documents held in Guildford. Moreover BAT had, in the 1980s, expounded a policy of seeking new opportunities in markets hitherto the preserve of state-owned tobacco monopolies383 and through the 1990s it successfully established a manufacturing presence in Russia, Ukraine and Uzbekistan and attempted to do so elsewhere in the region, most notably in Moldova. It was therefore suspected that the Guilford depository would provide a useful resource for studying TTC activities in the FSU.

As my searches for relevant documents started in 2000, the research for this thesis was unable to substantially benefit from subsequent endeavours to improve access to the Guildford documents by placing them on the world wide web. However, for some chapters supplementary searches of the new online resource were undertaken subsequently.

Approach taken Analysis of tobacco industry documents is a relatively new research technique in public health so methods are evolving.374 For the purposes of this thesis, I chose to use a sociohistorical approach, based on archival research methodologies.384 In light of the nature of the document collection being used, I complemented this approach with the more specific but closely related methodology of company documentation analysis.385

191

A year after I commenced the document work for this thesis, I was awarded, along with other colleagues at LSHTM, a grant from the US National Institutes of Health for a project titled“Globalisation, the tobacco industry and policy influence”. The methodology I developed for my thesis has therefore been shared by colleagues working on the larger project and the database I developed, described below, has been central to the organisation and analysis of documents for the larger project.

The idiosyncrasies of the Guildford Archive Although much of the archival research methodology aptly described by Hill384 is relevant, it is worth noting that the BAT archive in Guildford is not in some senses a true archive, which conveys an implication of a repository designed to enable information to be retrieved. As one archivist requested to review the Guildford Depository pointed out: “Its contents are stored and listed randomly, without any reference to their provenance. The database catalogue, whilst appearing to give relatively detailed information about the contents of the depository, is unreliable and obscures more than it discloses. In particular, there is no description of the contents of the files.”386 s

A database with a rudimentary index is provided to assist searching (Box 5-1). Unlike most archives, and certainly unlike the other tobacco company archives, the Guildford collection is not indexed at a document level, but at file level. Each file may contain anything from one to several hundred documents. Thus identifying relevant documents is a time consuming and roundabout process, requiring a broad search strategy.

Other complexities of searching the tobacco document archives are common to all archives and include the lack of a “controlled vocabulary” (the list of the subject headings under which documents in a collection are indexed) and spelling mistakes in titles and names. Spelling mistakes appear to be particularly common in documents relating to non-English language countries and differing translations from non-Roman alphabet languages were in my, and others’ experiences, a particular problem.377 Thus different words, abbreviations and spellings need to be used for the same term in order to help ensure all relevant documents are identified.357 Finally, it is sometimes difficult

s BAT of course did not want to make it easy for researchers to use its archive and thus the archiving system is far from helpful. However, over time, as explained below, some system is discernable.

192

to distinguish a planned action from one that was actually carried out or to determine whether planned policies were ultimately implemented.

Box 5-1 The Guildford archive Information about the contents of the archive is held in a Concordance database with the following fields: • G Box no. (Guildford box number or reference) • File Number • First page (Bates number) • Last page (Bates number) • Title (this can be the original title of the file, an assigned title given at the time of copying e.g. “Contents of XY’s desk” or a title given by the BAT lawyers) • Start date • End date • Owner (senior person responsible for originating and using the file) • User (person/s adding to the contents of the file; frequently the owner and user are the same person) • Projects (names of projects covered in the file; although the project names will not necessarily appear in the contents of the file) • Reports (a database term, meaning unclear ) • Responsive (numbers allocated to the court questions that the file was used in response to) • Index (a database term, meaning unclear, usually blank)

Understanding the provenance and weaknesses of the archive The first stage in planning an archival search strategy is to understand the provenance of the archive. This helps guide the researcher to unearth relevant material and interpret it more successfully.384 As noted above, the Guildford document collection was created as part of the discovery process in American litigation, although why BAT chose to submit such a broad range of documents that were not germane to the case has remained unclear. Possible explanations include avoidance of the high costs of legal review to identify what was covered; perhaps because BAT hoped to “swamp” the lawyers; or simply because it had lost track of what was and was not acceptable behaviour.387 The company had certainly not expected to have to make the collection publicly available and thus the collection is believed to be more complete than those of the American companies that were also party to the litigation. Nevertheless, it is important to bear in mind the documents that are not in the archive. I believe these fall into a number of categories: (1) documents that might have been (often innocently) discarded by BAT

193

staff in the course of their work, (2) documents from BAT’s many subsidiaries outside the UK that that were not retrieved as part of the original discovery process, (3) documents that BAT has genuinely or falsely claimed legal privilege on (see above), and most importantly, given evidence that the company and its lawyers has destroyed and altered documents, (4) documents BAT and its lawyers might have deliberately altered or destroyed.

It is also important to note the timescale of the documents. Due to the timing of litigation the documents held in Guildford date until 1996 at the latest, with subsequent BAT documents available only in Minnesota.

Planning and undertaking the search strategy For this thesis, the aim was to unearth documents on BAT’s conduct in the FSU, focusing on its political and economic rather than its scientific or research activities. Given the very crude nature of the database available at Guildford, a broad and iterative search strategy was used.

A background literature review, an internet search for relevant articles in regional English language newspapers including the Moscow Times, and a hand search of tobacco industry journals dating from the start of 1990 to the end of 2000 held at the British Library were used to provide background information including names of key players, factories, regions of interest etc that could then be used as search terms.

At the Guildford archive, country and regional names including abbreviations (e.g. FSU, CIS, NIS) were the first search terms used. From the documents thus identified and from the background literature review, other search terms were then identified. The most useful search terms were those relevant to BAT; thus names of BAT staff active in the region, names of factories BAT wished to acquire, project names assigned by the company to projects undertaken in the region and so on. In undertaking the searches, what Hill refers to as a “master name file” (a list of related individuals, organisations, including acronyms etc),384 and a list of search terms used were compiled. Relevant files were ordered from the depository, with file numbers being recorded. When new search terms began to identify the same files repeatedly, or became so obscure that they identified no new files, it was concluded that the searching was complete.

194

Searching took place over a two year period between July 2000 and July 2002. In total 193 search terms were used specifically for the former Soviet Union or Central Asia, with an additional 47 terms used to find related material that referred more specifically to Central and Eastern Europe. Different spellings were used for the same terms to ensure that all relevant files were identified. For example for the bank, Schroders I used the terms “Schroders, Schroeders, Shroders” and the catch all “Schr*”. 1834 files were retrieved from the FSU searches and 788 from the CEE searches (many of these files were essentially duplicates). From the dates of ordering, these took up to two years to be delivered by BAT. A few additional searches were used to identify generic BAT strategy documents where these were needed to contextualise the country-specific documents and strategies.

Organising the archival data In order to facilitate the comprehensive indexing and systematic retrieval and use of documents and related data, I constructed a relational access database. In this task I worked closely with an external consultant. The key fields in the database are outlined in Box 5-2 and these were decided on having worked with the documents for some time. Importantly the database, as well as allowing a comprehensive index of each document to be made, also allows profiles to be developed for each named person and named organisation, thus allowing the researcher to develop a historical profile of each individuals areas of activity and to identify the activities of an organisation. It also allows the researcher to develop a list of acronyms used in the documents. It then allows the selection and sorting of documents; most important for this analysis is the selection by country or region and the sorting by date. The database thereby facilitated the organisation of data in temperospatial sequence to construct a chronology of events and by networks and cohorts to allow analysis of the nature and extent of networks, political support and organisational affiliations. 384

195

Box 5-2 List of key fields in Access document database (document indexing section) 1. Doc ID (Unique ID no.) 2. File Location (location of file once arrived at LSHTM) 3. Source (usually the Guildford archive but also enables other sources eg Philip Morris website to be used) 4. Visit date (date of visit to the Guildford Depository or of internet search for other documents) 5. Reviewer (initials of researcher indexing the document) 6. File No. 7. Box No. 8. Start Bates 9. End Bates 10. Document Date 11. Title 12. Summary (a summary of document content) 13. Quotes (a list of useful quotes from the document) 14. Country (the country(s) to which the document refers) 15. Region (the region(s) to which the document refers. These are allocated automatically for the countries entered using the NLM system) 16. Subject (a drop down list of subject terms specifically developed for the project) 17. Co./ Instit’n producing the document ( the company that produced the document, usually BAT but other possibilities include Philip Morris, Japan Tobacco, RJ Reynolds, Tobacco Institute, Imperial Tobacco, Reemtsma, Liggett & Meyers, Council for Tobacco Research) 18. Author (document author) 19. Corporate Author (where no individual author is identified, a corporate author might be listed) 20. Recipient 21. Corporate Recipient 22. Cc (list of those the document is copied to) 23. Corporate cc (list of organisations the document is copied to) 24. Named persons 25. Named organizations 26. Brands (cigarette brands named in the document) 27. Projects (projects named in the document) 28. Notes (a space for other data) 29. Reference (an output of how a reference to the document should appear)

196

The files obtained through the above searches were then read through, either in at the Guildford depository or once they had been delivered to LSHTM. Relevant documents from within these files were then indexed in stages within the database using the above fields as described further below. A total of 808 documents were indexed under the geographical regions FSU or Central Asia or Eastern Europe. 356 focused on the FSU alone, 81 on CEE alone, and 7 on Central Asia alone, with 364 focusing on a combination of these areas.

Interpreting the data Analysis of the data is subject to a number of potential pitfalls; steps were taken to minimise or eliminate these.384,385 Two key issues are the risks of discounting or misinterpreting data. These are of particular concern in a value-laden subject area such as this, where the number of documents is so large that it is necessary to select a sub-set of them for analysis, and there is likely to be a tendency to make more of the evidence that supports our beliefs and less of evidence that contradicts them.388 This can be overcome to some extent by looking for both corroborating and conflicting evidence and taking some account of the quality and amount of evidence supporting a particular finding.

Steps recommended to reduce pitfalls in interpreting documents that were taken include:384,385

• Using an iterative process: documents were re-read at different times in order to avoid falsely rejecting documents that initially were not recognised as important and to avoid misinterpreting documents that had already been selected for analysis. • Continually asking what might be missing and remaining aware of the possibility that documents could have been produced to mislead, and recognising that the messages conveyed might be highly subjective.t • Attempting to validate the documentary data, contextualising and triangulating it with other documents and other data sources, including material available in the

t For example, staff might overplay their role in an event in order to seek praise or might misunderstand a situation.

197

published media (scientific and commercial) in order to provide corroborating or conflicting evidence. Where interpretation was particularly complex a second reviewer was used. • Attempting to interpret the documents in their own context, i.e. by the standards of the time and place the document was written rather than the present. This was achieved, wherever possible, with the assistance of local contacts. In interpreting company documents, Foster recommends a hermeneutic process involving seven stages which incorporate the validation techniques outlined above. Table 5-1 outlines these stages and how I have used them in the study of BAT documents.

198

Table 5-1 The hermeneutic process in analysing company documents Stage Process (and in brackets, specific process evolved for BAT document analysis) Understanding the Read and index individual documents searching for meanings of individual meaning. Reconsider in light of other documents. (Read texts through files and index key documents, using basic indexing terms. Re-read files and documents at a later date using accumulated body of knowledge to check the initial selection of documents for indexing and the indexing of key documents. At this stage add document summaries and start to allocate subject codings ) Identifying sub-themes Immerse oneself in the documents, begin to identify themes and sub-themes within each document and groups of documents (by reading through many relevant documents, identify emerging themes and allocate subject codings to documents) Identifying thematic Identify if there is a higher order theme. Start to cluster clusters documents along these themes. (ensure all documents given a “subject” coding in the database. Then order documents by subject and geographical codings, often in date order within the database. Paste document details into a word document under a country or theme and in date order). Triangulating Triangulate with other documents and other data documentary data Employing reliability and Could use second researcher. (Re-read over time as validity checks knowledge increases. Where uncertainty exists, use second reviewer. Ensure other documents and data provide corroborating evidence. Consider the circumstances in which the document was produced and the person producing the document) (Re)contextualising Place the documents in a broader context using other data documentary data sources (Ensure documents are interpreted in their appropriate geographical/cultural context at the time they were written using newspaper reports from the time and interviews with those working in this area at the time) Using representative case Select document subsets which are to be used for case material studies. Source: adapted from Foster.385

Thus, the files were read through and key documents indexed using the basic indexing terms outlined above. Once many documents had been entered in the database they were then revisited, with subject terms allocated and accurate document summaries added. This enabled me to review the documents for a second time and for the indexing to benefit from the knowledge accumulated in the interim. It also allowed research themes

199

to develop and thereafter document clusters could be identified within the database along subject themes or by country or region. Documents could then be extracted within these clusters or themes, and ordered, where useful, by another database category. Most often I used the date of the document to develop a temperospatial sequence. I then pasted document summaries into a word file to develop a “story” over time. Various materials were then used to triangulate and contextualise the data. First, other documents were used to ensure that the data had been correctly interpreted and accurately represented corporate strategy or conduct. Other materials were also used, most notably internet searches, newspaper reports, industry journal reports, conference proceedings and discussions or interviews with contacts in the field.

Case studies Document research, especially in an area where those producing the documents have sought to conceal what they were discussing, is by necessity iterative: while the broad objective (to examine the tobacco industry’s activities in the FSU at the time of transition) was specified in advance there was inevitably an element of serendipity with leads followed as they arose.384 Thus a detailed plan of investigation could not be specified in advance. Instead, from the documents identified in the wide-ranging document search strategy and indexing procedure outlined above, it was possible to identify research themes that could most easily and fruitfully be pursued. For example it soon became apparent from the documents that BAT, having initially investigated the possibility of investing in all FSU countries, narrowed its options. It had, for example, been extremely active in Uzbekistan and it was apparent that following BAT’s activities there would provide useful insights into its activities in other countries in transition. Thus, the hermeneutic process set out above was followed to generate relevant case studies which serve to illustrate BAT’s activities and behaviour and are presented in the following chapters.

Weaknesses A number of weaknesses that are largely intrinsic to tobacco document work must be considered. Searches at Guildford, whilst thorough, cannot be systematic for a number of reasons. Most notable is that the indexing of documents at file rather than document level makes it almost impossible to retrieve all relevant documents. Second, document

200

destruction and claims of privilege make it impossible to review all material. Third, a few documents are in Russian. For the latter, consultation with Russian speakers confirmed that these were identical to documents already obtained in English. Nevertheless by searching until the same files were repeatedly obtained or search terms became so obscure they failed to identify any files, I can be sure than my searching was as thorough as possible and certainly more thorough than that of much other tobacco document based work.

Selection bias is also a potential issue.353 Firstly only certain documents made it into the depository; secondly, only certain documents would be selected as part of the search process; and thirdly only some will make it into the case studies. The second and third issues here were to some extent overcome by studying a large geographical area in detail as this enabled me to undertake far reaching searches and to build up a broad grasp of the documents on the region so that I knew the case studies were representative of BAT’s activities in general.

A further pitfall of document work is to believe everything written in the documents and to assume that planned actions are carried out. For example, it is possible for someone seeking to impress his superiors to over-play his/her role in or the success of an event. The advantage of working exclusively with the BAT documents is that by immersing oneself in the documents it is possible to build up a knowledge and understanding of the BAT staff, their position and idiosyncrasies and thus their reliability as a witness. The documents can then be interpreted in light of this knowledge. On the issue of whether actions are carried out, as others have noted, it is not always obvious where the line between contemplated and actual action lies.353

This was overcome to some extent by researching a defined geographic area in detail which enabled me to obtain documents that backed up other documents, thereby testifying to their veracity. Wherever possible I also checked with “real-world” information sources that actions were carried out. For example that legislation was passed. In a few instances, most notably in relation to Uzbekistan, due to the political situation, it was difficult to obtain detailed information although I was ultimately able to check the varacity of key findings (see Uzbekistan case studies Chapters 9, 10 and 11).

201

The other main danger of document work that relates to a different cultural setting to the researcher’s own is that findings may be misinterpreted. Attempts to contextualise the data were made by wide-ranging reading not only on the subject in question but on the broader social political context at that time and by working with contacts in the field. In some instances these contacts were either involved directly in the work - for example in the case study on Moldova where I worked with native speakers of both Moldova’s languages (Russian and Romanian) to retrieve and interpret documents and related data. In other instances contacts were interviewed in person or by phone or asked to review my interpretation of documents or draft case studies.

The following six chapters are based on analysis of tobacco industry documents. Chapters 6 and 7 examine how the TTCs entered the newly opened markets of the FSU. Chapter 8 examines the situation in Moldova whilst Chapters 9 to 11 focus on Uzbekistan.

202

CHAPTER 6 How the transnational tobacco industry entered the former Soviet Union. Part I: Establishing cigarette imports

INTRODUCTION In their 1848 Communist Party Manifesto, Marx and Engels could have been forecasting the tobacco industry’s rapid entry to new markets almost a century and a half later. They wrote that when domestic capitalism ceased to progress or experienced a crisis, industrialists would respond “by the conquest of new markets, and by the more thorough exploitation of the old ones”.389 What they would not have predicted is the collapse of the Soviet Union in September 1991, the resulting upheaval, including widespread economic breakdown30 and the stampede amongst the transnational tobacco companies (TTCs) to gain a share of these previously closed markets.242

The TTCs saw the former communist bloc – then the world’s second largest cigarette market, with the Soviet Union alone the third largest market after China and the United States,390 as a golden opportunity.391 Patrick Sheehy, Chairman of BAT Industries between 1982 and 1995, reflected their attitude when he stated in October 1990 that “[t]he dramatic increase in the proportion of the world’s cigarette market now open to free enterprise [make these] the most exciting times I have seen in the tobacco industry in the last 40 years”. 383

British American Tobacco (BAT), the world’s second largest tobacco company,381 had in the 1980s, as outlined in Chapter 1, set out to harness opportunities in markets hitherto the preserve of state-owned tobacco monopolies.383 The importance of the former Soviet Union (FSU) in this policy is perhaps best summed up by Tony Johnson, a BAT board member and regional lead for Russia and central Asia. In an in-house publication, the BAT Bulletin, he described the opportunities there as “almost limitless”, explaining that:

“The emerging markets of Central Asia and the former Soviet Union in particular have immense potential and are of crucial significance to BAT.

203

As the long established markets of north America and Europe mature and contract - and they will continue to do so over the next five to ten years - it is vital that we find new markets to grow and expand our business. ………. the real opportunities for growth lie in the former Soviet Union and this is where we will be focusing much of our attention over the next few years.”392 Having recognised the opportunities available 393 the “[a]ggressive exploitation of the emerging markets in Eastern Europe”394 became central to BAT’s expansionist ideas and the need for “firm and aggressive strategies and plans to attack” specific markets was outlined.395

This chapter and the next, based on analysis of internal tobacco industry documents held at the BAT depository in Guildford, aim to analyse the reasons for tobacco industry’s interest in the markets of the former Soviet Union and its responses to their opening. This chapter focuses on the strategies that BAT adopted to increase imports whilst the next chapter focuses on how it established a manufacturing presence in the region.

Based largely on the TTCs previous entry to low and middle income countries, it has been suggested that TTCs follow certain steps in pushing for the denationalisation of state monopolies15,16,92 - an initial focus on licensed production before joint ventures are established, the use of intensive marketing with which the national companies cannot compete and the use of smuggling as a market softening technique. However, such work was unable to benefit from the unique insights provided by the tobacco industry documents as they were not available at the time. Moreover, little is yet known about the strategies and tactics used to enter the former Soviet Union (FSU). This knowledge is of particular importance for the few countries in the region where the industry remains state owned, and those in other parts of the world still without TTC investment.

METHODS The methodology employed in obtaining and analysing the documents was described in detail in Chapter 5. For this part of the study, the BAT depository in Guildford, UK was searched manually on-site using a broad and iterative approach. For this and the following Chapter over 600 documents were studied in detail and indexed in the

204

database described in Chapter 5, which enabled the sorting of documents by date and topic to construct a historical and thematic narrative. Supplementary data used to triangulate and confirm document findings were obtained from tobacco industry journals, BAT annual reports, newspapers, routine data and other published reports plus face to face and telephone interviews with local tobacco control experts.

RESULTS Reasons for BATs interest in the FSU BAT first became interested in the countries of Central and Eastern Europe (CEE) and the FSU in the late 1980s. With the assistance of reports funded by the British Department of Trade and Industry and Treasury, they were able to track developments including legislative changes and the Russian privatisation process.396,397,398 A BAT study undertaken to explore the way in which Eastern Europe could be opened up for the company noted that:

“If the changes in society and economy which have been announced are in fact implemented, the preconditions in these countries could be better than in many of the underdeveloped countries in the Third World.” 399 The enormous size of the former Soviet market made it especially attractive - as INFOTAB (an industry-run international centre on smoking issues formed in 1979) noted in 1990, the population of Eastern Europe, estimated at 426 million (two thirds of which was in the Soviet Union) was considerably greater than that in each of the European Community, the United States and Japan.400 BAT collated background data on the region which, inter alia, indicated very high rates of population growth and the young population structure of the central Asian republics.401 Market forecasts based in part on rates of population growth402,403 suggested these features made these countries particularly attractive. In addition, BAT noted the potential to expand sales to women who, until then, had low smoking rates404 and, like all young people, were more likely to smoke international filter brands. 323,404,405,406,

Other reasons for BAT’s interest were more specific to the region. The first was the shortage of cigarettes in the USSR. The observation that demand far outstripped supply was made first in Russia407, then Ukraine, where between 1987 and 1991 the supply of

205

filter cigarettes had halved from 30.8 to 15.8 billion,408 and later in central Asia.409 BAT assumed that consumption in these countries could be pushed back up towards its previous levels, even to levels found in Poland and Hungary, then amongst the highest in the world, as long as it was not allowed to stagnate at the low level that had resulted through supply shortages.395,410 As Anton van Waay, (Senior Manager in BAT’s New Business Development unit with an ambassadorial role in prospecting new markets), commented on the Ukrainian market:

“The estimate of 80 bns [billions] market demand seems conservative if compared with consumption per capita in Hungary and Poland. If this market demand is not satisfied over a longer period, it should be questioned whether this theoretical demand will not disappear?”395 Other factors that influenced BAT’s interest were the potential for increasing leaf production in the region to supply its new investments411 and fulfil the company’s aim of achieving growth through expanding leaf exports412 and, as I shall describe later, the FSU’s proximity to China.

Preliminary steps In 1989, BAT began to explore ways of entering markets within the Soviet bloc.413 It sought guidance from other western firms already successful in the Soviet Union, such as Rank Xerox, and was advised “not to prematurely jump into some grand joint venture” but to first “get “known” in the country”.413 In light of this advice, and following visits to the region in 1988 and 1989,414 BAT developed a cautious, step-wise approach to penetrating the market413 (see Box 6-1). Similar steps were outlined elsewhere399 and involved two main phases: the first to increase imports of BAT brands and the second to increase local production, initially via licensing and later joint venture.399,413

Once the Soviet Union collapsed, plans continued as before. BAT’s inceptive aim was to increase its share of the import market and the overall scale of imports415 as this offered the best short-term prospects416 and, as BAT had previously noted, could be achieved by “avoiding any significant investment on B.A.T’s part.”413 In the medium to long term BAT aimed to establish local manufacturing although the possibility of doing so

206

independently,416 not just jointly with local interests,413 was now recognised. In the early 1990s however, BAT became concerned about the risks and costs of establishing a joint venture417 or greenfield operation418 at a time of great political uncertainty and that the imminent dissolution of the Union would limit access to the whole Soviet market.417 It therefore focused its efforts on obtaining a share of imports.419 These efforts form the core of this chapter.

Box 6-1 BAT’s stepwise approach to penetrating the market413 1. Increase B.A.T.’s sales into the Beriozka [foreign currency] shops to give chosen brands more visibility in the Soviet Union through the ‘leakage’ which takes place into the local markets. 2. Reach a technical agreement with the selected Soviet parties. We estimate that, with our help, the U.S.S.R. cigarette industry could reduce its foreign exchange requirements for wrapping materials by around $10 million p.a. Part of the agreement would be that a proportion (say 60%) of this could be reserved for the import of BAT Group products for the domestic market. 3. Continue to maintain contact, particularly with the chosen factories and to provide them with redundant machinery and spares from B.A.T. sources in order to get the Group better known and trusted. 4. Export chosen B.A.T. brands into the Soviet Union. [using the foreign exchange released via (2) above]………… 5. Transfer from Exports to Contract Manufacture of a B.A.T. Brand in the Soviet Union under our technical supervision in order to expand B.A.T.’s share and brand franchise…... 6. An alternative to 5) above would be the development of a new brand within the Soviet Union using B.A.T. technology.” ….[Notes there are risks with this] 7. If by this stage confidence is appropriate, it would be possible to envisage licensed manufacture and/or some joint venture requiring further investment.

Establishing BATs brands in the market During the Soviet period, the supply of imported cigarettes was strictly controlled. BAT cigarettes had only formally been sold in airport duty free shops while BAT’s competitors’ (Philip Morris, RJ Reynolds and Reemtsma) brands had been sold in the Beriozka (foreign currency) shops, whence ‘leakage’ into the local markets secured brand visibility.413 These shops, established for tourists some 20 years previously, had, in BAT’s words, become “the shop window for western lifestyle” and enabled cigarette companies to make their products known, their desirability fuelled by the fact that locals had little access.399

207

These efforts served to establish a degree of brand recognition among locals and an aspiration to obtain these elusive western products, a situation that BAT saw as a major opportunity.322,420 BAT calculated that sales of western cigarettes would increase, not only by virtue of their better taste and quality, but because:

“western cigarettes are seen as relatively inexpensive status symbols. Anyone who smokes foreign cigarettes distinguishes himself from the egalitarian doctrine of socialism and thus demonstrates more individuality or personal freedom on a small scale.”399 The priority therefore was to establish BAT’s brands quickly so as to achieve a leadership position. 407,415 This was to be achieved in all new markets where BAT felt there was a “pent up demand for imported International Brands”.415 Patrick Sheehy, Chairman of BAT Industries, repeatedly stressed the need to register BAT’s brands in as many countries as possible.421 He recommended that each company within the BAT Group review the issue monthly with their lawyers and continue “the practice of covering the use of brands on other, non-tobacco, products.”421 In the FSU this was a complicated process. The concept of trademark ownership differed from the west – trademarks, where owned, were owned by the state not individual producers, with brands produced collectively across a number of factories.422 Moreover, closure of the Soviet Trade Mark office on 1st February 1992 made the means of registration outside Russia unclear.421

Simply placing brands on the market was insufficient; an effective distribution system had to be combined with brand marketing or what BAT referred to as its “pull” strategy:

“To increase unit volume sales, BAT should combine its brand awareness effort (“pull” strategy) with limited investments in distribution and channel management strategies and opportunistic use of declining state distribution assets (“push” strategy)” [original in caps].324 Brand marketing was required to build loyalty for brands that would ultimately be manufactured in the region. Thus, establishing a brand strategy or portfolio was an immediate priority.423 Concerned at the potential advent of advertising restrictions,322

208

BAT, like the other TTCs, moved quickly to exploit the media opportunities available with massive advertising and sponsorship, 271,424,425,426 unknown in the Soviet era. 68,70,427

Certain consumers were seen as particularly important in establishing brand loyalty. Women were targeted through selected advertising323 and those living in urban areas through focused distribution systems.323,324 But above all the emphasis appears to have been on younger opinion leaders322,423 as one Russian marketing study suggests:

“Three factors are very clear: (i) Most young Russians aspire to western international F.M.C.G [fast moving consumer goods] brands and will forego “necessities” in order to afford them. (ii) Those that can afford to consistently buy western brands are younger consumers who are involved directly or indirectly in private enterprise and, ipso facto, are the “opinion leaders”. (iii) Consequently, advertising investment in brands now can establish a loyal core franchise on which to build a wider franchise as consumers’ disposable income rises.” 322 Concern about advertising restrictions also led BAT to consider the role its corporate image could play:

“[a] good, high profile, corporate image will assist in opening up new markets, being seen as a good venture partner, and supporting our relationships with governments. If advertising restrictions make it difficult to introduce new brands, we may need to rely more heavily on the BAT name and its good corporate image.”415 The desire to establish a brand presence was also based on the need to enhance company visibility in order to promote BAT’s position in joint venture negotiations.428 BAT felt that its lack of a globally recognised brand akin to Philip Morris’ Marlboro or RJ Reynolds’ Camel would be a disadvantage. For this reason it felt that “BAT should change company names to include BAT where there is an opportunity to do this, to improve visibility of the name”.429

209

Securing a place in the import market Taking on the competition From the outset, BAT realised that time was of the essence stating “...Mikhail Gorbachev’s amended statement: “If you come too late, history will punish you” also applies to us.”399 Yet BAT was more cautious than its competitors in initial attempts to enter the market, most notably in establishing legal imports; in its own words, BAT “has not appeared to be as quick off the mark” [original emphasis].429

In the rush to establish brands in the market, BAT’s competitors were at an advantage as they had already completed several of the steps identified by BAT413 and outlined in Box 6-1 – their cigarettes were already being sold in the foreign currency shops413 and Philip Morris had established a licensing arrangement for Marlboro in the 1970s and developed trademarks with Soviet colleagues including the Apollo Soyuz cigarette to commemorate the space mission.430,431

In 1990 BAT’s major competitors again edged ahead - RJ Reynolds and Philip Morris concluded deals with the state importing agencies for Russia (Rosvneshtorg) and the Soviet Union (Prodintorg)417,432 for the import of 22-23 billion Philip Morris cigarettes and 19 billion Reynold’s cigarettes417,430 to Russia and yearly supplies of 25 billion Philip Morris cigarettes thereafter.430 Media reports of the deal,433 including details that these companies airlifted 34 billion cigarettes to the region,74 suggest the contracts were honoured. Indeed, Philip Morris’ 1990 deal was estimated to have added US$50-100 million to the company’s operating profit in 1991433 despite BAT’s assertion that low prices were paid.417,419

BAT was keen to secure a similar deal419,432 but its contacts with state organisations occurred too late for it to be included in these initial orders.417 Although later documents suggest it finally negotiated the official import of 12.5 billion cigarettes to Russia,421 its competitors’ brands clearly had a head start. Thus by the time BAT was negotiating imports, other companies had begun to secure licensed production arrangements.417,419

In exchange for import orders, the FSU countries wanted assistance in modernising their tobacco industries419 so national authorities were likely to give preferential treatment to imports from companies interested in investing.416 It appears that Philip Morris’ success

210

in securing import orders stemmed from its ability to convince governments of its longer term investment interest, even where none existed, by signing a letter of intent “well ahead of any serious commitment to an investment.”416 BAT took a similar approach in preparing its strategy for securing import orders with state agencies.434

Funding imports In the final years of the Soviet Union, the state had so little cash that it paid for the large orders placed with Philip Morris and RJ Reynolds partly by barter transactions: third parties in the west would buy wood, oil and fertilizers from the Soviets and in turn make payments to the TTCs.417 Once the Soviet Union had collapsed, the economic position deteriorated further with a major economic crash in 1992. BAT recognised that there was no cash to pay for orders and that rapid price increases might lead to a decline in sales429 yet it remained so keen to establish imports that it was largely prepared to ignore these risks.

Instead BAT took advantage of its global reach to develop three main systems for ensuring the FSU countries could fund cigarette imports – counter-trade, smuggling and the use of money from aid packages to fund cigarette imports. At one of the early meetings on market entry strategies attended by Patrick Sheehy (Chairman of BAT), Martin Broughton (later Chief Executive and Chairman of BAT) and Ulrich Herter (Managing director Tobacco, BAT Industries) it was noted that:

“A key determinant of the ability to buy cigarettes will be the availability of hard currency either through aid packages or, in the case of the Russian trading organisations, through exports of oil or other commodities sold in world markets. In addition, there could also be possibilities for counter-trade (e.g. Brazilian leaf for oil)”416 At subsequent high level meetings further details emerged on how exports to Russia would be funded by aid packages. Minutes of a December 1991 meeting, for example, state: “[i]t was noted that Hollywood [a BAT brand] would be included in the offer for the EC [European Commission] financed project but would need to be manufactured in Europe, using a Souza Cruz blend of leaf imported from Brazil.”407 Minutes of the subsequent meeting in May 1992 note, in relation to a contract to export 12.5 billion cigarettes to Russia via the state importing agency, that :

211

“[t]he major problem, which is being experienced by all importers, is arranging finance. Priority is being given to arranging payment from funds available from aid programmes, including a special sale of Jockey Club from Argentina, financed by funds from an Argentinean aid programme.”421 International assistance programmes were also seen as offering less direct benefits such as funding to train local managers435 or improve leaf production.419

Although the documents suggest that European Union funds were used to pay for cigarette imports, I have been unable to verify this. The main source of European aid money to the region is the TACIS Programme, a technical assistance programme established in 1991 to support the transition to democracy and market economy in Central and Eastern Europe and the former Soviet Union. In the first eight years of operating it committed a total of €4,226 million of funding to projects.436 Although private sector development is a priority area for the programme, EC officials contract projects out to consultants and believe it would be difficult for money to have been misspent in this way (Personal correspondence Per Eklund, European Commission).

As in the late Soviet era, the use of barter transactions was another key strategy. Despite their shortfalls, BAT had used barter deals to import its HB brand into the Soviet Union noting that “it [barter] has to be utilized whenever possible and economically expedient in order to maintain or improve the brand awareness of our products”.399 BAT learnt more about counter-trade from transnationals in other sectors, most notably Pepsi Cola International which had been running a franchise in the FSU for almost 14 years and had “wide experience of barter dealing as means of payment”.437 To secure imports into Uzbekistan for example, Murray Marr (BAT’s project leader in central Asia) realised that BAT would have to devise “creative means of securing payment which is ultimately convertible to hard currency”428 and on meeting with the Chairman of Uzbakalea (the Uzbek state food distribution agency) to establish opportunities for “immediate importation” of BAT brands into Uzbekistan, agreed to produce a list of “acceptable barter goods”.438

Ever optimistic, BAT even saw some benefits in the shortage of hard currency, noting that it could bring advantages as “[c]igarettes were and are always an important hard

212

currency article in times of need.”399 The same point was made by the Financial Times which noted:

“In Russia’s hard pressed economy cigarettes are regarded not as a health hazard but as socially benign. Demand heavily outweighs supply, giving the status of a surrogate currency which, unlike the rouble, does not depreciate.” “Truck loads of Pall Mall, HB, Hollywood, Kent and Lucky Strike are now the currency in which many a domestic business deal is settled as well as a staple part of the retail trade.”439

The role of smuggling in accessing new markets Reports of the high level meetings on market entry referred to above show how, in their rush to establish imports, BAT planned “to pursue several channels in parallel”.416 While some documents mention a number of import channels and operators,407,417,420,421 the legality of which is difficult to establish, others suggest that many of these operators worked in illegal channels, where re-export was a key concern.420, 440 , 441 Overall it appears that smuggling was a key market entry strategy.

Previous work on BAT’s role in smuggling indicates how BAT staff use euphemisms such as “general trade” and “transit” to refer to smuggling in their documents, while clarifying that transit “is essentially the illegal import of brands from Hong kong, Singapore, Japan etc. upon which no duty has been paid.” 442,443,444 The BAT UK and Export Limited Company Plan 1993-1997 outlines how, in Central Europe, high import duties make the export environment unattractive but that “GT [General Trade] opportunities exist”.445 BAT industries 1993 draft Tobacco strategy document suggests that BAT was prepared to condone smuggling where excise policies were not favourable:

“We will seek to persuade governments to operate sensible excise and import policies, such that transit trade is reduced or eliminated, recognising that where there is an imbalance, market forces will operate. Rationale: Transit trade is volatile and disruptive to the orderly operation of markets. It is in BAT’s interest that markets are legal, taxed and controlled.

213

However our primary responsibility is to meet consumers’ demands as profitably as possible”.415 A regional marketing document confirms BAT’s willingness to smuggle when excise rates are high, stating, “GT supplies of IBs [international brands] will continue to be the dominant supply source until duties are reduced to the range 30-50% (currently 98% in Russia)”.446 BAT suggest that a gradual shift towards legal imports might occur in 1994, but its plan to “ensure that transition to fully paid duty scenario can be achieved without significant disruption to supply” implies that until that time, the majority of imports were smuggled.446

A more detailed document exploring strategies for establishing brands in the FSU includes the following assumptions:

– “GT will flourish” – “GT Market will formalise in 1993 and continue through the plan” – “Powerful International Brands will drive and benefit from the GT market”; and – “Establishment of brand image and consumer franchise are pre-requisites for success in GT.”420 How a market supplied by contraband would “formalise” without direct industry involvement is unclear. The document goes on to outline individual brand strategies, including for State Express 555 the aim “[t]o generate sufficient franchise to benefit GT”. BAT also notes its competitors’ use of smuggled cigarettes stating “[h]igh volume competition brands will exploit GT”420 and Philip Morris’s strategic trademarks were “likely to remain mainly GT imports until duty change”.446

A distribution study for Russia prepared for BAT by Bain Link (a management consulting firm) outlines the difficulties of operating in a chaotic and seemingly largely duty not paid market,447 but once again suggests that BAT was operating largely within this illegal sector:

“The [distribution] structure in place could serve as the skeleton for backward integration into [a] fully integrated distribution network when and if Russia becomes a duty-paid market.”447

214

One of the disadvantages of developing a vertically integrated distribution system (in which BAT would develop its own distribution infrastructure) was that BAT could then be held responsible for tax compliance, something they were clearly keen to avoid. This document, for example, notes that vertical integration at the importer stage “would involve taking responsibility for tax compliance, resulting in a considerable competitive disadvantage in the current environment.”447 Similarly, by integration at the distribution stage “one level above and below the importer, there could be some implied responsibility for tax compliance.”447

The company’s marketing strategy for the Russian Federation produced in 1994405 and related documents448 confirm how these issues play out in practice, stating:

“At the present time it is also not advisable for BAT to own or operate regional warehouses as this step would directly give us the burden of complying with customs regulations. Should the market become more oriented towards DP-cigarettes [Duty-Paid] we would have the chance of changing our distribution strategy based on our existing regional presence. We should also be clearly aware of the fact that our opportunities for volume expansion are dependent on our importers to increase their volumes with special customs treatment.”405 Other documents suggest that the majority of BAT’s exports to central Asia and the Caucasus were illegal, as was much of the company’s trade to the rest of the FSU.441 One document dated May 1994 states clearly “[i]t is known that business in this region is a transit business”.440 Another dated the same month but referring exclusively to Uzbekistan, where BAT was about to establish a joint venture with government queries “Would it be possible to negotiate the legal import and promotion of BAT brands now that UZBAT will manufacture in future? Do we want to do this?”, thereby implying that until that point imports were illegal.449

Importantly, despite the apparent imperviousness of the borders of the Soviet Union, cigarette smuggling was not new. BAT’s role in cigarette smuggling including its major smuggling operation into China has previously been documented.442,443, 450 Other

215

documents highlight how, in the Soviet era, as part this Chinese operation, BAT supplied the Sino-Soviet border trade with illegal cigarettes. This trade was already underway in 1986,451 and involved substantial volumes of cigarettes. Monthly market reports from BAT (Hong Kong) Ltd for 1989 suggest that approximately 36-54 million cigarettes were supplied for this border trade each month 452 , 453 with considerable increases predicted.453

While BAT’s Hong Kong subsidiary was keen to promote this trade, uncontrolled “leakage” of these cigarettes from China was causing problems for BAT.454,455,456 This led Wai Pong, the General Manager of BAT’s Hong Kong subsidiary to arrange for a consultant to report on the border trade. Visiting the “end buyer’s office” the consultant reported that:

“They [the Heilongjiang army] had signed a contract with Xiamen SEZ [Special Economic Zone]Trade Company Ltd and the latter had to supply 6000 cases of SE 555 [BAT brand State Express 555] to them per month. The Russian[s] liked SE 555 most but they also bartered other brands of cigarettes. The supply from Xiamen could not reach the contracted quantity in these two months and Xiamen claimed that the supply from BAT was not enough.”457 Following a series of internal meetings about this illegal trade454,455 a decision was reached to reduce the supply of smuggled cigarettes to that area in an attempt to control the leakage:

- “quantities supplied to Xiamen will be a factor of train loads, rather than a factor of container loads. We will do our best to ensure that nothing is left over for Mr Xie to either sell locally, or collect and sell elsewhere.”456 However, instead of stopping the trade, more direct links were later established with the Chinese Army along the border. 451,456 as:

“..we have never doubted the existence of a demand on the Sino-Russian border. As a result of one important new circumstance, i.e. meeting and talking direct to the Army, we believe, and we hope we are right, that the business is genuine. In that belief it would be wrong to shut it out, or at least not try to and prove conclusively the validity of that belief.”456

216

Indeed, one of the major attractions of the FSU was its border with China, which after the opening-up of the FSU remained the ultimate prize as the world’s largest but officially closed cigarette market. Thus BAT’s plans for the FSU included strategies for ensuring the illegal supply of cigarettes into China across this border. For example the 1994 “CIS [Commonwealth of Independent States] Operating Plan” outlined how BAT aimed to “[e]stablish supply [of the brand State Express 555] to cities on [the] Chinese border where personal transit opportunities exist.”446 A 1994 meeting exploring BAT’s approach in central Asia implies that BAT intended to formalise this supply route when it notes that “[t]he strategy for this brand [State Express 555] will be reviewed when more is known about the GT [general trade] routes to China”.458 The document goes on to outline how this specific issue would be examined in a study to be commissioned on distribution.458

DISCUSSION The documents analysed in this paper provide insights into the strategic thinking behind the TTC’s focus on the FSU markets and highlight the elaborate and apparently devious and illegal tactics used to establish imports to the region. Although based on BAT documents, comments within these documents and my brief review of other company documents suggest that the attitude and practices of BAT’s competitors were similar459,460 and included the use of smuggling.430,420,446

Marketing, brand and corporate identity The documents suggest that the FSU was a market ripe for exploitation, with the potential to provide immediate returns on investment. Favourable features included the undersupply of the market,407,408,409 the vast population402 and in some areas its young age structure,401 high male smoking rates, and although few women smoked,118,128,404 they, along with the young of both sexes, were more likely to smoke international filter brands. 322,404,405,323 As indicated in the documents, the TTCs set about exploiting these conditions, along with the status afforded to western products in general and cigarettes in particular399,322 by glamorising the western way of life,86 and, as in its previous entry into Asia,16 making concerted efforts to appeal to women, young people and opinion leaders.461

217

As outlined in Chapter 2 the levels of cigarette advertising that resulted were phenomenal. Advertising practices have in turn been reflected in smoking prevalence patterns, as I highlight in Chapter 4 and have shown previously in Ukraine and Belarus.129,130 The higher rates of smoking among women compared with previous surveys and far higher rates of smoking among younger than older women, suggest that female smoking is a relatively new phenomenon. This targeting of women, few of whom previously smoked, refutes the common industry claim that advertising is only used to encourage brand switching and not to recruit new smokers.

In establishing imports, determining a brand strategy was clearly paramount. The documents highlight how the company’s brands and corporate image were to be used synonymously to promote the company and market its products.415,429,428 The collective focus on brands and corporate image was used to pre-empt the negative impact of potential advertising restrictions. In this regard it is of note that the Chairman focused on the need for brand registration, in particular the requirement to register non-tobacco products,421 highlighting the important role of brand stretching. Collectively these documents highlight how corporate and product branding and marketing go hand in glove and show that marketing restrictions that focus solely on advertising without limiting the use of brand or company promotions will have a limited impact. This is illustrated by evidence from Brazil in which the response of BAT’s subsidiary, Souza Cruz, to a comprehensive advertising ban was to use the company logo in promotions.462 Overall the contrast between the Soviet era when branding and advertising were unknown concepts and the post-Soviet era where branding and marketing were essential, co-dependent and highly prevalent strategies, could not have been greater.

Smuggling Shepherd outlines how smuggling reaches its peak around the time that a TTC enters a market but was unable to implicate the TTCs directly in this process.15 The documents reviewed in this chapter highlight clearly the role that BAT played in smuggling. Smuggling offers a number of advantages to TTCs. It acts as a market softening technique by creating demand for the (often highly desirable) smuggled product before a domestic manufacturing presence is established. This undermines local firms (which can then be more easily and cheaply acquired), makes it easier to argue the need for local

218

manufacture on the basis that the demand for quality products led to the illegal supply and reduces government revenues.15,92 Smuggling also ensures a ready supply of cheap cigarettes, thereby encouraging consumption by undermining public health efforts to moderate demand through price controls.

Having conquered the FSU and central and eastern Europe, the TTCs major focus is the forbidden fruits of China. BAT’s carefully organised smuggling arrangements highlight the importance it places on ensuring a brand presence in such closed markets. It is clear that Sino-Soviet cigarette smuggling was just a small part of BAT’s major smuggling operation into China442 and that BAT’s interest in the FSU was predicated, at least in part, on the potential to manufacture cigarettes that could ultimately be supplied to the Chinese market.

In the FSU, the TTCs were undoubtedly aided in their smuggling efforts by the notoriously corrupt state customs committee 463 and its poorly paid officers. 464 As expected, the widespread avoidance of import duties had a clear impact on government treasuries at a time of major economic hardship.465 Moreover, smuggling was not the only way that duty-not-paid cigarettes entered Russia. Legal systems for duty-free cigarette imports were established under the auspices of the Russian Orthodox Church and the National Sports Fund.466,467 In the case of the orthodox church, cigarettes were registered as humanitarian aid (along with any good given free) and the Church was allowed to import them duty free.468 This bizarre system apparently originated when a Dutch cigarette producer approached the Orthodox Church with a request to give the church its surplus production. Although I found no industry documents directly linking any of the TTCs to the Orthodox Church, other sources suggest Turmans, Rothmans’ Dutch affiliate gave 250,000 packs of roll-your-own tobacco to the Soviet Union and thus may be the company that initiated this deal.7 The Daily Telegraph newspaper also reported that in the mid-1990s the Moscow Patriarchate entertained executives from Philip Morris.469

219

Outcomes Although there were clear risks in importing cigarettes to a region in such economic disarray and BAT was relatively slow to act, later documents show that BAT’s predictions of the potential of these markets was correct, noting that

“[f]or several years now, the best new business development opportunities have been in the newly opened up markets in Russia, Eastern Europe and Central Asia and in the opening and fast developing markets of the Far East.”470 This is perhaps more an indication that the conditions were ripe for industry expansion rather than a reflection of BAT’s skill in harnessing the opportunities. The latent demand for cigarettes and desire for change after decades of Soviet rule, combined with the emergence of newly established governments whose focus was by necessity on state- building and embryonic legislatures that lacked any tobacco control measures, meant that failure was almost impossible, especially when such high consumption rates had already been achieved without the benefit of marketing.

Although progress was not always straightforward, particularly in Ukraine where trading conditions were difficult following the 1998 economic collapse,287,288 the east European markets have generated good profits. By 2001 the five markets of Russia, Poland, Hungary, Romania and Ukraine, where 10 years previously BAT had no production facilities, accounted for 12% of Group volume.381 Moreover, the 2001 annual report stresses that the group’s success in Eastern Europe has given BAT confidence in its ability to “grow share by entering new markets”. These trends are reflected in industry- wide data for the Eastern European region. Despite the difficulty of estimating consumption accurately due to the high rates of smuggling, since 1992 consumption in eastern Europe has grown steadily compared with a decline in western Europe.317 It is also the region with the greatest growth in market value – a 21.9% increase between 1995 and 1997.317

Summary The TTCs were were quick to exploit the chaos that emerged from the FSU, although many others – the oligarchs, the mafia, corrupt officials and even the Orthodox church - also stood to gain. It is clear that governments in markets that the TTCs seek to exploit

220

need to protect themselves and their citizens from industry tactics. Most importantly they need to protect themselves from smuggling if they do not want to lose enormous revenues to the TTCs and encourage increased consumption of cheap cigarettes. They also need to ensure adequate legislation is in place to prevent unregulated tobacco promotions and thereby protect their populations from the impact that entry of these companies can have on smoking patterns and hence the health of these vulnerable populations.

221

CHAPTER 7 How the transnational tobacco industry entered the former Soviet Union. Part II – establishing a manufacturing presence

BACKGROUND The previous chapter established how, based on initial exploratory visits to the region413,417 and advice from western firms already active in the area, BAT had outlined a cautious, step-wise approach to market entry that could be divided into two main phases – first to establish imports and second a manufacturing presence. This chapter focuses on the second of these phases.

Despite BAT’s initial concerns about the risks and costs of establishing a joint venture or green field operation,417,418 it soon recognised that it could not build a significant market presence without establishing domestic manufacturing facilities.418 This chapter explores how BAT responded to the new opportunities for investment in cigarette manufacturing in the FSU, how it prioritised these opportunities and above all, the tactics it used to establish a manufacturing presence. It aims to give a brief overview of these tactics; subsequent chapters, most notably Chapters 8 and 9, will explore the tactics used in individual markets in greater detail.

Before exploring these issues, a brief review of the market reforms underway at that time is warranted. As set out in detail in Chapter 1, Section 1.4, fierce debates centred on the speed of reform30 and led to the emergence of two schools of thought, the “shock therapy” approach of rapid and extensive privatisation and the more “gradualist” approach wherein the creation of a competitive environment and necessary institutional infrastructure were seen as essential precursors to privatisation.37 Shock therapy won the day30 and in less than a decade the number of firms privatised across the former communist bloc was almost 10 times greater than the number privatised in the whole of the rest of the world in the previous 10 years.29 Although some successes have been seen in central and eastern Europe, the consequences for the FSU were often disastrous and while some have prospered, life under “capitalism” has, for many, been worse than under Communism.30,36

222

The tobacco industry was just one of the many sectors that were privatised. In Russia for example the state monopoly was abolished in 1992 and state tobacco firms were transformed into independent joint stock companies that were then privatised, enabling the transnational tobacco companies (TTCs) to acquire their assets.471 A similar model applied in other republics, although subject to differences in the overall pace and extent market reforms,35 leading to major differences in tobacco industry structure (Chapter 2).

METHODS The methodology for obtaining and analysing documents is described in detail in Chapter 5. Supplementary data used to triangulate and confirm the document findings used in this chapter were obtained from tobacco industry journals, BAT annual reports, newspapers, routine data and other published reports plus interviews with local tobacco control experts.

RESULTS Restructuring BAT Group operations to maximise new business opportunities Investment in new markets, particularly those in the Eastern bloc countries, was such a priority within BAT that it led to the creation of a new, dedicated unit and a major company restructuring. At a series of high level meetings in the early 1990s Patrick Sheehy (Chairman, BAT Industries), Ulrich Herter (Managing Director Tobacco, BAT Industries) and others416,472,421 agreed to establish a New Business Development unit (NBD) to maximise BAT’s opportunities in emerging new markets. The unit was described as “a central team to co-ordinate the identification and assessment of investment opportunities, to prepare proposals and to be responsible for the negotiations leading up to investments in eastern Europe and the USSR.”416 Although the unit’s final remit extended beyond this region, particularly to the Far East,412 identifying opportunities for investment in the FSU was the immediate priority.416

The NBD unit was actively supported by Sheehy and Herter,429 and given high status within the company, transferring to BAT’s London headquarters in February 1992 because Sheehy “perceived that greater impetus would be created in grasping new

223

business opportunities if the function reported centrally”.473 Yet the unit alone was insufficient to deal with the many opportunities available at that time474,475 and BAT, concerned that its decentralised structure was a disadvantage in developing new business opportunities compared with its competitor Philip Morris,473,415 underwent further changes.473 The most significant was a major company restructuring, in January 1993, entitled “Project Rubicon”. Two of BAT’s operating groups, BAT Co. and BATUKE (BAT United Kingdom and Export), combined their management capacity and restructured along regional lines, creating four regional business units that divided the world market among them. It was felt that closer coordination between domestic and export marketing operations would enhance the ability of BATCo “to exploit the growing opportunities for its brands in world markets”.476

Prioritising investments within the region Compared to Philip Morris, BAT had little experience in the FSU.430 This is reflected in recurrent misspellings of country and politician’s names,477,478,421 and attempts to obtain advice from other large multinationals.478,437 Inexperience, combined with the rapid political changes at the time, resulted in chaotic initial attempts to prioritise countries for investment. The earliest ranking listed East Germany, Hungary, Czechoslovakia, Yugoslavia, Poland, Bulgaria, and Romania, followed by the then USSR.399 By September 1991, the USSR had collapsed and priorities changed. Hungary was identified as the “best short-term prospect for investment” and Russia, Ukraine and Moldova as showing “the most promising potential”.416 The relative priority given to countries continued to vary472 until 1992 when a clearer system was established and priority markets were identified as those which were:

– “of sufficient size; – potentially profitable; – likely to be open for investment within two to three years; – not already committed to a competitor” 412. This led to a more systematic list of 39 first and second priority countries, 23 of which were in the former communist bloc, with all 15 FSU countries listed as either first or second priority.412,479 In order of appearance on BAT’s “first priority” list were Ukraine, Russia, Poland and Macedonia (by this time BAT had acquired the Pécs factory in

224

Hungary).412 Thereafter the order changed slightly, although usually for obvious reasons. For example, as trouble flared in the Balkans, Macedonia and Serbia disappeared from the list,479 while Uzbekistan was prioritised474 as a result of a review of central Asia.409

Prioritisation was based largely on the existing and potential cigarette market size.409,412 Russia stood out by virtue of its sizable population (149 million) and the former Communist markets differed from others prioritised by BAT by their relatively high rates of consumption.479 This was especially so for Poland and Hungary395 which then had amongst the highest rates of tobacco consumption in the world. In central Asia the gap between existing cigarette production and estimates of potential demand may also have been a key factor - the countries prioritised had the widest gap as well as the largest potential demand.409 Beyond these main criteria, political and economic circumstances and geographical location were taken into account although they were rarely adverse enough to exclude a potential investment.409,393 For example, Schroders advised BAT against investment in Kyrgyzstan due in part to political instability,409 yet BAT still pursued this possibility.480 Similarly BAT (along with at least one other TTC) spent from early 1993 to the end of 1995 attempting to invest in the Grodno Tobacco Factory in Belarus, 481 , 482 when other investors realised that market liberalisation was most unlikely under President Lukashenka,483,484 who even BAT’s advisers had described as an “unbalanced farmer”.485

Issues of competition Although, as outlined in Chapter 6, BAT had established a staged approach for entering the former Soviet markets that allowed for a gradual increase in the level of investment and risk required, ultimately, once the communist system collapsed, they had to act far faster than anticipated.

The competition to acquire assets was intense. In Uzbekistan alone fourteen letters of intent were signed for the establishment of a joint venture with the Tashkent Tobacco Factory between 1991 and mid-1993.486 (See Chapter 9) Even by late 1993, as BAT was finalising a deal to acquire this factory, Uzbek government offices were being “deluged with letters from various consultants fronting for PMI [Philip Morris International], RJR [RJ Reynolds] and Rothmans.”487

225

BAT recognised Philip Morris as their most “aggressive” competitor470 but others were also a threat, particularly RJ Reynolds, Rothmans488 and Reemtsma, which according to BAT aimed to be the largest company in Europe and was “expanding aggressively in eastern Europe in pursuit of this goal.” 470

My analysis suggests pressure to acquire assets rapidly arose for three main reasons. First, there was a perceived window of opportunity in which the absence of marketing restrictions and low costs of advertising could be exploited in order to establish cigarette brands in new markets (See also Chapter 2).322 The second was the scale of competition between the TTCs; the third was the industry’s wish to avoid competitive tendering.

Competition to acquire factories inflated prices480 and BAT discovered that its competitors were prepared to pay considerably more than it, thereby exacerbating concerns about the threat of competitive tendering:489

“BAT’s experience in recent months shows that Philip Morris is prepared to offer substantially higher prices in situations where BAT and Philip Morris compete to acquire a controlling interest in a monopoly cigarette supplier in the former eastern bloc (“FSU”). Specifically, BAT believes it was outbid by a factor of up to some three times in both Klaipeda in Lithuania, where Philip Morris was the successful bidder, and Almaty in Kazakhstan, where Philip Morris has made the winning offer in a competitive tender ….……. Over the last 3 years a number of deals have been structured involving the exchange of western skills and finance for ownership interests in FSU and Asian factories. Initially, these deals were very often achieved as the product of bilateral negotiations with local partners who where (sic) often perceived as having a very uncertain future in the absence of aid. In each case, control passed to the western company. The deals BAT struck in Hungary and Ukraine typified this approach. As the number of deals has increased, it has become evident that western investors are prepared to place a significant value or “premium” on participation in the emerging markets. It now appears that a trend is developing

226

towards the local factory (or its owner, the state) adopting a more demanding stance and putting the opportunity out to competitive tender. …….. While BAT may have failed to outbid Philip Morris in two cases, this could merely reflect a preparedness of Philip Morris’ part to overpay for what they consider “strategic” markets”490 Notably, despite the intense competition between TTCs to acquire assets, this rivalry rarely resulted in open tenders and the anticipated benefits for the seller, but led instead to closed deals wherein the TTCs could gain the advantage. BAT went out of its way to avoid a competitive process,489 doing so successfully in Ukraine490 and Uzbekistan480,,491 and attempting the same in Russia492 and initially in Moldova (see also Chapters 8 and 9).493 Other TTCs were playing the same game.494 In Belarus, where BAT attempted unsuccessfully to acquire the Grodno tobacco factory, it deliberately sought to confuse the Belarussians about the size of its offer, which was under half that offered by Reynolds:

“BAT may be forced to uncloak its offer which has hitherto been deliberately supported by ‘smoke and mirrors’. BAT’s offer actually values Grodno at $22.3 million (less than half RJR’s offer).” 495 The TTC’s desire to establish monopoly positions321 undoubtedly exaggerated the pressure to acquire assets rapidly. So too did the regional approach to investment, in which marketing and manufacturing strategies were to be more tightly coordinated on a regional basis415 and an individual country was seen as providing a manufacturing base for a whole region. Thus BAT saw Moldova as the site from which to export to the Caucasus (Armenia, Azerbaijan and Georgia) and Uzbekistan as BAT’s centre in central Asia423,496 and “a key market in our global brand strategy that is trying to link the Far East with Europe.”497 This approach presumably was aided by the TTC’s apparent willingness to facilitate cigarette smuggling into countries where they did not have production facilities. (Chapter 2)

In many instances the TTCs successfully established monopoly positions even in populous countries (Chapter 2). Traditionally, the absence of competition in monopoly markets results in the absence of advertising, with potential benefits for tobacco control. Here however, the documents suggest that even where BAT anticipated establishing a

227

monopoly position, its marketing plans were developed on the basis of a competitive market due to the perceived threat from imports and desire to establish brands before possible advertising restrictions.423

Tactics Having noted that “Eastern Europe cannot be seen as a monolithic, political and economic block… [but that]…[e]very market must be looked at and analysed individually and the relevant strategy for opening it up be developed and implemented individually”,399 BAT devised a set of tactics to gain entry. These had been identified early as part of the NBD strategy where it was noted that:

“If BAT Industries is to be successful in entering new markets it must:

– make early contacts in the priority countries; – be quick to respond and be flexible; – build up the trust of the relevant governments by advising over industry structure and excise; – make agreements with governments that are transparently fair if the they are to survive changes in regimes; – be prepared to commit to investment; – be prepared to offer licensed manufacture of our International Brands in order to build up local awareness of the brand supported by advertising; – offer leaf and manufacturing know-how; – be prepared to barter if necessary, especially leaf; – win the confidence of local manager by explaining BAT’s decentralised culture, offering to upgrade facilities, explaining how improved sales and marketing expertise and the introduction of selected international brands, can benefit the company, and explaining management incentives.”474

Contacts At one of the earliest meetings on new business development it was noted that the first priority for the NBD team would be to travel widely in the region and establish contacts at the highest level.416 It would be assisted in this task by consultants “who had good connections with the leading industry and political figures in the country”472 and

228

accredited representatives who would “maintain contacts with national authorities”.474 BAT’s use of political contacts was absolutely key and so extensive that it cannot be covered in detail in the space available here. Rather, I will focus on the way in which BAT sought to “build up the trust of the relevant governments” firstly by selling the benefits of privatisation and transnational investments in general and secondly in promoting BAT versus its rivals.

Selling the benefits of privatisation BAT had remarked in relation to Poland that “[t]he difficult stage is to get the business privatised”419 While most post-Soviet governments were receptive to privatisation, others, most notably in BAT’s context those of Belarus and Moldova, were resistant. BAT therefore set about using a number of tactics to persuade such governments of the benefits.

In this they were supported by the international financial institutions. The IMF for example assisted by pushing generally for privatisation in Belarus, as BAT’s advisers noted:

“Lukashenko [President of Belarus] initially put a stop to the privatisation process, but re-started it in March under pressure from the IMF, which was withholding credit until the Government showed more commitment to economic reform.”498 The same was seen in Moldova when government failure to approve tobacco industry privatisation led the IMF to suspend its Extended Fund Facility disbursements in 1999 (see Chapter 8).

In relation to Belarus, BAT argued in documentation prepared for the Deputy Prime Minister Ling’s visit to the UK, that there was an:

“urgent need for investment in Tobacco Factory Grodno to modernise and expand production facilities, to improve quality and quantity of cigarettes for supply to the Belarusian market and to replace imports coming into the country without duty being paid. The increase in production at Grodno and the replacement of imports by domestic production will result in Government revenues being increased.”499

229

It also emphasised the “increasing fiscal and excise revenues to the Belarusian Government” as one of the benefits of BAT’s investment.500 BAT maintained this stance despite documents suggesting the company was itself responsible for smuggling cigarettes to the region (see Chapter 6) Moreover, as outlined in Chapter 3 BAT continues to market its brands heavily in Belarus271 despite having little official market share, a behaviour that can only adequately be explained by it having a substantial share of the illegal market, estimated at some 40% of the total.211 Other tactics used to facilitate privatisation of the Grodno factory included the use of government insiders to push for privatisation and help ensure the replacement of a resistant factory leader with one more receptive to BAT’s approaches.501,502,503,504,505

BAT also wanted to invest in Moldova (see Chapter 8). A series of slides and figures perhaps produced to sell a BAT joint venture option to the Moldovans reviews three options for the tobacco industry. It suggests that the single joint venture option (as opposed to no foreign investment or dismantling of the monopoly with privatisation of individual units) would maximise foreign income and government revenue, 506 but ignores the effective monopoly power that this would give BAT.

To further persuade the Moldovan government of the potential benefits of BAT’s investment, the Minister of Agriculture and others were invited by Sir Patrick Sheehy to visit Brazil to demonstrate that, despite the industry being privately owned, the Brazilian government still derived substantial revenue from it (further details are given in Chapter 8).507

Economic arguments and excise advise Economic arguments were also used in the less resistant countries. Notes on Patrick Sheehy’s visit to Uzbekistan and meeting with President Karimov outline how Sheehy emphasised that increased quality and exports of tobacco leaf plus import substitution could increase government revenues:

“SPS [Sir Patrick Sheehy] mentioned to President benefit that would accrue to state viz: $20-25 mn value of leaf (improved quality and exports) $20-25 mn import substitution

230

others i.e. at least $50 mn p.a.”508 A hand written note next to this memo also indicates that Sheehy was keen to stress the macro-economic benefit of investment. Such claims were repeated publicly509 and made elsewhere, for example in Ukraine, where Sheehy, again in direct contact with the President, Leonid Kravchuk, emphasised the importance of tobacco taxation revenue and suggested that, with BAT’s input, leaf imports could be reduced.510

Plans to use such economic arguments had been developed at one of the early meetings that established BAT’s strategy for new markets where it was decided that:

“Mr Herter would set up a small team independent of New Business Development Department, who would offer a service to governments, indicating the revenue earning possibilities of tobacco and other products and advising on the optimum structure for excise regimes. Discussions which the Chairman had with Mr. Attali and Mr. Freeman at EBRD [the President and Vice President respectively of the European Bank of Reconstruction and Development] suggested that funds could be available to cover the cost of specific projects undertaken by the team.”421 The European Bank for Reconstruction and Development (EBRD) was established in 1991 to nurture a new private sector and use investment to help build market economies in the region. Since its inception it has had a policy of not loaning for tobacco.511 BAT record that EBRD staff approached the company in October 1991 to enquire whether the bank “could help BAT to fund projects in Eastern Europe”.512 This led to plans for the Chairman to contact EBRD to establish whether the bank could provide assistance with financing472 and resulted in the discussions detailed above. According to EBRD however, BAT misinterpreted their approach which was aimed at the broader group of companies then subsumed within the BAT Industries portfolio (at the time BAT was divesting its retail assets and concentrating on tobacco and financial services).383 Although there can be no doubt that BAT would have seen EBRD as a valuable ally, there is no record of EBRD providing assistance or funding to BAT (personal correspondence Beverley Harrison and Chris Beauman, EBRD, 2003).

231

Whilst BAT sought to promote its excise advice as a service to prospective business partners,508,513,514 other documents make clear that the advice was proffered largely for BATs benefit.515,408,423 As outlined in the 1993 draft BAT Industries Strategy:

“Priority will be given to initiating dialogues with governments on excise in order to establish appropriate excise structures compatible with the structure of our new product ranges. Rationale: The structure of excise can greatly affect the structure and profitability of our markets. BAT has wide expertise which it can use to advise governments on how to achieve their objectives while not distorting cigarette markets and at the same time improving our competitive position.”415

Selling BAT In both the resistant and receptive countries, BAT used a number of tactics to sell itself as preferable to its competitors as a joint venture partner. A simple tactic was to arrange for key players to visit BAT’s established ventures such as the Pécs factory in Hungary, its first joint venture in the region, to persuade them that “BAT are good and welcome foreign partners.” 516 Other tactics included the promotion of BAT’s image as a responsible partner and selling BAT’s strengths.

The corporate image The previous chapter indicated how BAT thought that the promotion of its corporate image would assist in opening up new markets.415 This ploy was used in a number of ways.

In 1993, just as the competition for assets in the FSU escalated, the World Tobacco Symposium was held in Moscow and used as a venue to promote BAT’s image. In his keynote speech, Ulrich Herter attempted to sell BAT’s role as a tax collector along the lines described above:

“Not only do we collect excise but we also advise and set up schemes in many countries, that will deliver to their governments the revenues they need in a predictable and orderly way.” The speech then went on to promote BAT’s apparent altruism:

“[F]or me the most satisfying area is sponsorship of the arts.

232

Our company has recently brought a young Russian musician to London to perform in a classical concert and, on a larger scale, next month we are bringing The Art of Holy Russia exhibition to London’s Victoria and Albert museum.”517 Other superficially altruistic actions were undertaken in Moldova at the point when BAT was bidding against Reemtsma to acquire assets there (see Chapter 8). 518 , 519 , 520 Elsewhere however, as notes on a visit to St Petersburg make clear, BATs assistance (like that of Philip Morris) was predicated on direct advantages to the company:

“PM [Philip Morris] has been looking at a project involving the renovation of the Moscow City Centre (Boulevard Ring) including bus shelters, kiosks, lighting systems etc. This project is being developed by an Italian architect with connections to the Raucci family who are exclusive exporters to Russia of Marlboro (26 million per day). The project is expected to cost $20 m and is likely to be rewarded with 49 years advertising in the city centre. Although I believe PM could well be looking at such a project nobody would be foolish enough to expect a guarantee of anything in Russia particularly for 49 years. I suggest we explore the avenue of assisting the Moscow government with infrastructure improvements only if we can derive benefit in the short term. For example during JV negotiations.”521 [emphasis added] Selling BATs strengths BAT also went out of its way to sell its perceived strengths in relation to other TTCs. The strengths identified included:

– “first rate experience in managing local brands; – good range of established international brands; – long experience of dealing with governments of every hue; – ability to give advice on industry and excise structure; – leaf expertise, including advice on leaf growing and improving quality, and supply of seed; – wide experience in managing local factories; – manufacturing expertise and availability of second hand machinery; – ability to barter, especially for leaf;

233

– commitment to management development including inter-company transfers.”412 Offers of assistance in these areas were used in varying degrees and according to the needs of the host country in order to promote BAT as a potential partner. In most instances it was clearly to BAT’s advantage to provide these services in any joint venture it established. During negotiations to establish a joint venture with the Prilucky factory in Ukraine, for example,522 BAT offered to supply “know-how, management skills, marketing and distribution skills, technology, access to trade marks etc.” 523 Similar offers were made in Belarus and Uzbekistan,524,525 with the addition in Belarus of guarantees of employment, sponsorship of the local community500 and legal support in an ongoing dispute.503 In some instances more devious offers were made. For example in Uzbekistan BAT planned to contract the Tashkent factory to manufacture an extra 5 billion cigarettes per year using BAT-owned secondary equipment in order “to lock the Uzbek authorities into an agreement to negotiate a JV [joint venture] exclusively with BAT by the simple expedient of demonstrating quickly a genuine commitment to assist the Tashkent factory” (see Chapter 9).525

Two key areas of expertise: tobacco leaf and local brands These general offers of assistance, including the excise advice outlined above tended to also be offered by BAT’s competitors. However, BAT differentiated itself from its competitors through two more unique areas of expertise as a letter from Sheehy to President Nazerbayev of Kazakhstan illustrates:

“Our proposals will certainly differ from those of our competitors in two key respects. First, unlike those competitors, we have considerable experience in developing domestic leaf crops; in Brazil alone over 220,000 tonnes of tobacco are grown by farmers under contract to BAT every year. No other company can match our record in this area. Secondly, whilst our proposals feature the development of international brands at the Alma-Ata cigarette factory, we do not believe that all smokers in Kazakhstan will necessarily either be able to afford, or indeed will wish to smoke, these cigarettes. A significant element of our proposal therefore features the development of existing national brands at affordable prices.”526

234

BAT’s greater emphasis on local brands527 was a key strategy differentiating itself from the competition526,528 and one seen as attractive in the FSU.527 It occurred for a number of reasons – BAT’s lack of a truly global brand on a par with those of its competitors, its historical emphasis on local brands381 and most importantly because of the greater profit margins gained from these brands in the early years of investment.421,515 BAT’s experience from their first acquisition in the region, the Pécs factory in Hungary, had shown that “the upgrading of local brands offers the best immediate prospects for significant increases in market share and profits.”421 This was particularly the case in times of economic hardship. Philip Morris, who may have learnt the hard way, having been left with a warehouse full of unsold Marlboro cigarettes in Russia 529 , 530 also adopted this policy.531

BAT’s favourite sales pitch was, however, its tobacco leaf expertise. As noted in the 1993 draft Tobacco Strategy: “High quality Leaf, and the provision of advice to farmers, is one of BAT’s key competitive strengths, particularly when negotiating positions in new markets.”415 BAT went out of its way to use this asset as a letter from RHL Taylor, a leaf expert within the BAT’s NBD Unit, to BATs subsidiaries illustrates:

“In presenting BAT to potential partner, NBD uses a range of material emphasising the company’s strengths. One obvious, and very attractive, advantage we have over our competitors is our involvement in leaf which has proved to be a critical success factor in many of our negotiations.”532,533,534

BAT used this expertise to its advantage by emphasising the potential for import substitution,508,510 offering assistance in leaf production and implementing leaf trials sometimes, it appears, purely as a public relations exercise.535,536 As noted above, it arranged for Ministry of Agriculture representatives to visit Brazil to witness the leaf produced under contract to BAT first hand.507, 537 BAT was also keen stress the uniqueness of these skills in relation to other multinationals510, 538 pointing out, for example, to members of the Uzbek Presidential delegation to the UK in November 1993 “PMI’s lack of leaf development expertise” and that “RJR had none of BAT’s unique leaf development expertise”.539

235

These arguments successfully won allies in Ministries of Agriculture528,537 who, keen to enhance their leaf production,540 were then used to support BAT’s case. In Kazakhstan where BAT entered the bidding war very late, Van Waay noted that “[f]ollowing the visit by the Chairman, the Ministry of Agriculture appears to be pro-BAT and is keen for BAT assistance in tobacco production.”537 Interestingly, Philip Morris’s Kazakh proposal, which was ultimately successful in the competitive tender (not least, as referred to above, because Philip Morris offered three times more than BAT), also attempted to emphasise its leaf interests by naming Universal Leaf as a partner although according to BAT:

“PM’s [Philip Morris] proposal shows that its authors do not take their own “agricultural programme” seriously. Universal Leaf (ULT) is identified as PM’s “partner” at the beginning of the proposal... A local leaf development effort is promised, ….[but]… Thereafter, ULT is never mentioned in the proposal. On the contrary, PM intends to introduce an American blended cigarette (Marlboro) under license, using “imported cut tobacco and materials”….. Therefore PM does not even intend to cut the tobacco, much less grow it, in Kazakhstan” 541

In Uzbekistan, despite the promises made to President Karimov of a potential $25million earning from tobacco leaf508, BAT privately admitted that “[t]he potential leaf export business is fragile at best”497 and that it would be difficult to bring the leaf crop up to international standards.486 Nevertheless, the opportunities for BAT were good because “[d]omestic leaf can be used as a strategic source for the FSU.”486

The relative ignorance of the FSU governments in relation to tobacco leaf was apparent. And it is ironic that BAT continuously referred to and offered trips to Brazil where Christian Aid have since highlighted the appalling plight of tobacco farmers under contract to BAT’s subsidiary Souza Cruz.542

236

DISCUSSION A number of issues become apparent from this overview of BAT’s efforts to establish a manufacturing presence in the FSU. BAT’s willingness to undertake major organisational change illustrates the priority it attached to investing in this region. The other TTCs also prioritised the region and this led to considerable pressure to acquire assets rapidly. Most notable however are the tactics BAT used to push for privatisation and secure investment. These included offering inappropriate economic advice to encourage privatisation and excise advice that disadvantaged governments whilst benefiting BAT, generating confusion over pricing, avoiding competitive tendering and using other anti-competitive means such as the supply of secondary equipment in Uzbekistan to lock governments into deals. BAT promoted its “good” corporate image and its perceived strengths, most notably its tobacco leaf expertise, to sell itself as a suitable business partner. Its task was facilitated by the international financial organisations’ support for privatisation and the naivety and inexperience of post-Soviet governments,484,25 which BAT exploited. As notable as the industry’s focus on the economic impacts of tobacco was the total absence of any debate about tobacco’s health impacts. Although the industry is usually quick to document and counteract any opposing or “anti” activity 543 , 544 and detailed the region’s fledgling “anti-smoking movement” in one of its earliest reviews,400 through the remaining documents reviewed I found details of only one international tobacco control meeting, held in Poland in November 1990 and attended by delegates from the Soviet Union.545 Thus, as far as I can tell, the governments of the newly independent states received little if any effective tobacco control advice or appropriate information on the true health and economic impacts of tobacco.

BAT and its competitors The previous chapter showed how BAT missed the first set of state import orders while its competitors established licensing arrangements first. Here the documents indicate that BAT lacked experience in the region, was reluctant to pay market prices and was outbid at competitive tenders. Ultimately its competitors were quicker to establish a manufacturing presence as they had been in obtaining import orders. RJ Reynolds was the first to invest in Russia and Ukraine, in the latter, despite BAT’s belief that it could

237

be first into the market.408 Philip Morris also acquired assets in Kazakhstan before BAT had established their monopoly in neighbouring Uzbekistan. This suggests a surprising torpidity or caution on BAT’s behalf as, despite the risks involved in investing in the FSU, the odds were stacked in favour of the TTCs: with high potential rates of consumption, a large gap between supply and demand, the support of the international financial organisations for privatisation and little if any countervailing support for tobacco control, it was surely difficult for the TTCs to fail. Indeed to my knowledge (see Chapter 2), there is only one significant investment in the region from which the TTCs have withdrawn – RJ Reynold’s initial investment in Azerbaijan.231

Market reform The evidence presented in this chapter lends weight to criticisms of the IMF’s rapid reform policy by highlighting the ease and extent to which the TTCs exploited the lack of regulatory frameworks and absence of a competitive environment. They were able to avoid competitive tenders, hoodwink governments over prices, establish monopoly positions, avoid import duties through smuggling and influence taxation systems. This lack of regulation also posed risks to the investing companies but BAT was largely aware of and seemingly happy to dismiss these and related problems. 546 , 547 , 548 ,437 Moreover, the literature suggests that whilst the international financial organisations considered the risks to international investors, they failed, at the time, to consider the detrimental impact that the behaviour of the oligarchs, corrupt government officials and international investors have had on the reform process (this is discussed further in Chapter 10).30,29,37 In retrospect this reveals a degree of naïvety about the propensity of international investors to avoid competition and ignore established standards of business practice. 549

Economic arguments Just as the IMF had promised economic salvation from market reform, so too did the TTCs. Most prominent amongst BAT’s tactics to encourage denationalisation and TTC investment was the promotion of flawed economic arguments that merit more detailed review. BAT’s overall strategy was to stress the macroeconomic benefits of the development of an indigenous tobacco industry.421,499,508 Put simply, these arguments are a mirror image of the economic “myths” the industry more frequently espouses to

238

oppose the implementation of tobacco control policies. 550 In practice, the complex economic impacts of tobacco remain a topic of some debate. But it is becoming increasingly clear that the economies of only very few countries, those with large tobacco leaf export sectors, depend to any great extent on tobacco551 and that, despite the early deaths of smokers, smoking seems to impose a net economic burden on a country.552 In the FSU, only Kyrgyzstan has traditionally relied to any great extent on tobacco leaf for export earnings553 and although, as shown in Chapter 3, its tobacco leaf production has now recovered from a fall in the mid-1990s, tobacco leaf exports have declined and imports increased since the early 1990s.554,555

The strength and plausibility of the economic arguments BAT has used stem from the grains of truth that lie within. Yet it was the industry that, directly or indirectly, often prevented any potential economic benefits from accruing. The arguments fall into three main areas. The first is that improved cigarette quality will help reduce illegal imports.499 Given that the TTCs use smuggling to pressure for market opening and were already smuggling cigarettes into the FSU (Chapter 6)242 this argument was at best disingenuous. The second focused on the excise earning potential of tobacco,421 alleging that BAT investment would increase excise revenues.500 Whilst excise can of course be raised from tobacco (or indeed any other good), TTC investment would not per se increase excise revenues over those that could accrue via a state owned industry. Instead, given the TTCs greater propensity to lobby for excise reductions, the opposite may occur16 as witnessed in Ukraine556 and as we shall see in Chapter 10, in Uzbekistan and undoubtedly elsewhere. A drop in excise obviously helps the TTCs increase sales, undermines tobacco control and further reduces government revenues already depleted through cigarette smuggling.

The third economic argument was that import substitution of both tobacco leaf and cigarettes would lead to financial gains.510,508,499 The cigarette substitution argument is again clouded by the TTCs willingness to illegally import cigarettes when tax policies were unfavourable (Chapter 6). Nevertheless, across the region the predicted import substitution of cigarettes has occurred to the extent that the considerable increase in cigarette imports seen from 1990 to 1996 has now stabilised. However, due to the TTC’s successful creation of demand, the vast increases in annual cigarette production,

239

estimated at approximately 200 billion cigarettes across the region between 1991 and 2000, have been channelled directly into local consumption and no trade surplus has yet to emerge (Chapter 3).

The suggestion that leaf imports could be substituted in countries such as Ukraine510 that had not traditionally been tobacco leaf producers seems farcical given the subsequent marked fall in leaf output and increase in imports.555,556 Indeed local experts have alleged that tobacco production collapsed as a direct result of TTC investment.556 BAT had noted privately, even in Kazakhstan and Uzbekistan, areas better suited to leaf production, that production of quality leaf would be difficult.486,497,557 And it appears that in Uzbekistan BAT saw the development of domestic leaf supplies as a means of providing cheap supplies for its own use in the region486 rather than the international revenue-earning export it had promised Karimov.508 And it was surely clear that, whatever BAT’s brand strategyu, the entry of other TTCs to the region with their more exclusive focus on international brands317 would lead to a requirement for Virginia leaf rather than the oriental leaf traditionally produced in the FSU. BAT indicated as much in its comments on Philip Morris’s proposal for Kazakhstan.497

The extent to which BAT misled governments on this issue is illustrated by the data presented in Chapter 3 which show that leaf imports have increased exponentially in the FSU particularly in countries that have received tobacco industry investments leading to a spiralling trade deficit. They also show that leaf production remains lower now than ever previously recorded. In addition, TTC investment has also led to an increase in various other imports including machinery, cigarette paper and filters, further influencing the trade balance.556 In Uzbekistan it is rumoured that BAT’s failure to deliver on its promise to help tobacco farmers and improve tobacco leaf has finally and recently led to government anger.558

The experience in the Czech Republic, where Philip Morris commissioned a much derided report claiming that early deaths from tobacco could bring economic benefits shows how far along the line of flawed economic arguments the industry is prepared to

u BAT has a more diverse brand portfolio than most other TTCs and tends to include local brands amongst this.

240

go.559,560 Yet these economic arguments are certain to have carried great weight in countries in such economic disarray.

Corporate image Finally BAT’s promotion of its corporate image merits comment. The previous chapter outlined how the company’s image was to be used to promote the company and help market its products, in part to pre-empt the negative impact of potential advertising restrictions. This chapter highlights the true rationale for BAT’s social actions, whether sponsorship of the arts or donations to emergency relief funds: in the short term to promote BAT as a business partner and in the longer term to help sell cigarettes should marketing restrictions reduce access to advertising and increase reliance on corporate image. This highlights again the potential misuse of corporate social responsibility.561

Summary Overall this chapter highlights the considerable priority attached to the former Soviet markets and the disingenuous tactics the TTCs use when negotiating entry to new markets. It highlights the dangers of a rapid transition from socialist to market economies when the supporting institutional structures are not in place and the investing companies use business practices that fall short of international standards. Governments must be aware of the tactics highlighted in this chapter, wary of flawed economic arguments pedalled by the tobacco industry, and better informed of the true economic and health impacts of tobacco in order to make more informed decisions about the future and structure of their tobacco industries. The international community has a responsibility to help inform governments and assist them in the development of tobacco control capacity and ensure that where privatisation occurs it is properly policed so that benefits to the state are maximized and harms minimised. Experience elsewhere suggests capacity development should occur both within government and civil society, although the latter will be more problematic in countries in this region where the transition to democracy has at best been incomplete. 484,25

The next four chapters look in more detail at events in two countries, Moldova and Uzbekistan.

241

CHAPTER 8 A review of transnational tobacco company plans for a privatised tobacco industry in Moldova

BACKGROUND Moldova, a small, agriculturally dependent state of 4.3 million people is now Europe’s poorest country. 562 Although Moldova initially followed the economic reforms recommended by the international financial organisations, as in much of the former Soviet Union, outcomes have been dismal.36,237,563

Moldova was the Soviet Union’s largest producer of wine, grapes and tobacco. It has the highest death rate from liver cirrhosis in Europe – over seven times the European Union (EU) average and although smoking rates there are lower than in much of the former Soviet Union (FSU) (see Chapter 4), its standardised male lung cancer death rates are also above the EU average.329 Nevertheless, the International Monetary Fund (IMF) pressured Moldova to privatise its tobacco and alcohol industries.10 Privatisation has, however, proved contentious, and the tobacco industry and much of the alcohol sector remain in state hands.

Privatisation of state owned tobacco industries raises several concerns. The IMF argues that governments will be more willing to support tobacco control when they no longer sell tobacco (personal correspondence Peter Heller, IMF, October 2005).107 However, economic theory suggests that privatisation, like trade liberalisation, will increase competition, driving down prices, increasing advertising, in turn stimulating consumption. Evidence shows that trade liberalisation leads to increased consumption. 5,98,107,108 And although rigorous empirical evaluations of privatisation are lacking, its alleged benefits (at least in a competitive market) - increased efficiency and reduced unit costs15 – are harms when the product is damaging to health.39 By contrast, inefficiencies of state owned monopolies can be beneficial to public health. Previous chapters show that TTC entry to the FSU was associated with a massive surge in advertising, a weakening of tobacco control and an increase in cigarette consumption, with greatest increases in countries receiving TTCs investments. Moreover the surveys in eight of the fifteen former Soviet countries outlined in Chapter 4 found that, after adjustment for socio-economic status, religion and urban dwelling, prevalence, particularly in women,

242

tends to be higher in countries with the strongest TTC presence; in Moldova for example, female smoking prevalence was the second lowest of all countries surveyed (3.9% vs 15.5% in Russia) and male smoking prevalence was the lowest (43.3% compared with 60.4% in Russia).

Despite concerns about the impact of tobacco industry privatisation,39 the IMF continues to promote it.10 This chapter seeks to inform the privatisation debate by reviewing internal tobacco industry documents on Moldova. This information will help inform efforts to minimise the impact of TTC entry on population health.

METHODS The methodology used to obtain and analyse the documents is described in detail in Chapter 5. In addition to and following the detailed searches performed in Guildford, additional web-based searches were performed for this Chapter. First, as the story was not specific to BAT, searches were also made on the Tobacco Documents Online (www.tobaccodocuments.org) and Legacy Tobacco Document Library (http://legacy.library.ucsf.edu) using the terms Moldova*, Chisinau, Kishinev (alternative spellings of the capital city), KTF (Kishinev Tobacco Factory) and Tutun. Alternative spellings for Moldova such as Modova, Moldavia and Moldovia were also used but identified only two additional documents. Second, in an effort to ensure the comprehensiveness of our initial search, a search of the BAT documents on the University of California website was made, but no additional documents were identified (this was prior to the launch of the GAP project – see Chapter 5 and thus only limited numbers of documents were available at this time). Third, as the story continues until the mid to late 1990s, a preliminary search of BAT documents dating mostly from 1996 to 2001 and made available in the Minnesota depository through more recent smoking and health litigation in the USA was also made. This identified a few additional documents. It should be noted however that there is no index or metadata for the BAT documents held in Minnesota, other than a list of Bates numbers and the only system of searching available is a manual search. As such, the search for these documents is unlikely to be thorough.

243

As the two main companies of interest were BAT and Reemtsma and Reemtsma was not party to the litigation described above, the analysis is based largely on BAT documents.

Additional information was sought from secondary sources including international and national newspapers in both national languages – Romanian and Russian and by working closely with tobacco control experts in Moldova and Romania.

RESULTS The saga of tobacco industry privatisation in Moldova. A joint venture agreement with the German manufacturer Reemtsma (now owned by UK based Imperial Tobacco) was first announced in 1994 but was never implemented. The Moldovan government then employed Arthur Anderson to value its tobacco business564 and in an international tender in 1996 Reemtsma outbid BAT to buy a 60% share. Its $59 million bid would have made it the country’s largest foreign investor,565 but in 1997 the new government annulled the deal.236,566,567

In February 1998, as part of a plan to develop the tobacco industry, the monopoly, Tutun SA, was split into eight separate companies - the Tutun cigarette factory (known variously as Tutun CTC, Kishinev or Chisinau Tobacco Factory) and its fermentation plant, plus 7 other fermentation plants. Under sustained pressure from the IMF568 the government once again moved to privatise the tobacco industry. Despite IMF threats to sever relations, parliament rejected the necessary bill in November 1999 and again in April 2000.569,570 This led the IMF to suspend its Extended Fund Facility disbursements and in turn to temporary suspension of World Bank structural adjustment credit disbursements (although loans continued in other sectors) and European loans.10,237,,571 Left with little choice, the new government finally approved the bill in October 2000.

Inadvertently IMF action also contributed to the collapse of the then centre right government. This led to further political instability (there were four prime-ministers between 1999 and 2004) and the ascendance of the Communist Party, its popularity fuelled by the failed reforms, thereby reducing the chances of successful privatisation.572

Although various components of the Moldovan tobacco industry were until recently listed on the government’s privatisation website573 and the IMF continues to push the

244

issue,574 there has been little subsequent action. Media reports suggest that failure to establish a government commission led to asset stripping as soon as the law on privatisation was passed.575 It is uncertain therefore whether the current impasse results from ideological opposition, reluctance to privatise industries seen as national assets, lack of interest in a degenerating industry or the benefits allegedly accruing to government officials involved in the monopoly. The press has labelled such officials the “tobacco mafia” 576 , 577 and reports that businesses belonging to senior politicians monopolise tobacco distribution from Tutun CTC.578 A further disincentive to privatise could arise from the reported deal, in November 2001, allowing Moldova to use tobacco assets to cover debts on Russian gas imports.579 Whatever the reasons, it is apparent that the industry is steadily disintegrating.

Reasons for BAT’s interest BAT was particularly interested in the Moldovan tobacco industry as it was a major leaf producer, had a cigarette factory reputed to be the best in the region (a legacy of Philip Morris’ earlier collaboration) and a major domestic and export business.407,419,423,580,581 As a result, in one of its earliest attempts to prioritise investment opportunities in the region, Moldova along with Russia and Ukraine were seen as providing the most promising potential.416

Privatisation BAT initially aimed “to gain exclusivity, or at least the first option in negotiations, should the Moldovan tobacco industry be privatised” 493 and a presentation outlines its preference for a single joint venture over privatisation of individual units.506 From 1992 onwards, BAT maintained an active presence in Moldova,582 prepared a development plan for the Moldovan tobacco industry583 and aimed ultimately to establish a joint venture including cigarette manufacture, leaf production and processing. Documents suggest that BAT attempted to gain favour in various ways. These included establishing leaf trials,493,584 considering the purchase of surplus Moldovan tobacco stock520 because it believed that “BAT’s ability to buy this tobacco will endorse our commitment to invest in that country”519 and by donating funds in aid of flood relief.518 A deal in which BAT would provide advance funding for tobacco leaf to enable the Moldovans to buy fuel for curing and transport was also negotiated585,586 and although it is unclear if this came to

245

fruition, BAT certainly provided Moldova with agricultural machinery.587 BAT also worked with Mr Cojacaru, Director and President of the Tobacco Association of Moldova, to lobby for privatisation of Tutun CTC588 and as was descrbed in previous chapters, invited the Minister of Agriculture to visit Brazil with the express aim of convincing him of the benefits of privatisation:

“The objective of the visit is to expose the Minister to a tobacco industry which is entirely in private hands and demonstrate that despite this the Brazilian government still derives substantial revenue. It is hoped that this may persuade the Minister to review his thinking on privatisation of cigarette manufacture in Moldova”507

Initially it appeared that BAT’s activities had secured an agreement to cooperate with the Moldovan industry, 589 but in 1994 their efforts were thwarted by Reemtsma,582 which suddenly emerged as the preferred joint venture partner.518 BAT allege that Reemtsma used a “notorious wheeler and dealer”, to ‘grease palms’ to secure the 1994 agreement, which later collapsed partly as a result of a press campaign instigated by BAT.582 Imperial Tobacco who now own Reemstma, when asked to comment on these allegations, claimed to be unable to do so as it only acquired Reemstma after the events in question.

The documents also suggest that on realising Reemtsma’s interest, BAT suddenly changed track. It gave up attempts to secure a closed deal, (which generally appears to be BAT’s preferred option and one successful elsewhere – see Chapters 7 and 10), pressing for a “full and fair tender” for which they thought they had obtained the support of the Speaker of the Parliament.518

The tender was duly announced,590 taking place in 1996. BAT believed their own offer was better and alleged again that the decision favouring Reemtsma was the result of political pressure or inducements to members of the tender committee.582 Stories along these lines emerged in the press at the time591 and to my knowledge were not publicly denied by Reemtsma. Imperial Tobacco was once again unable to comment (Smithson J, Imperial Tobacco, Personal communication August 2004). By this stage BAT was very concerned about Reemtsma’s dominance in neighbouring Ukraine and that, should

246

Reemtsma acquire the Moldovan industry, its position in Ukraine would be strengthened through the significant share Moldovan products held in the Ukrainian market. 592 With plans for a free-trade area between Ukraine and Moldova, BAT saw this as a major threat to its business in the region592 and again campaigned aggressively against the decision. Plans to use Moscow-based papers and the Sunday Telegraph to put pressure on the government by highlighting the dangers of investing in such a corrupt country appear to have been successful.593 BAT also began to explore ways to influence the situation more directly:

“Meetings held with influential members of parliament and Government to establish the facts and get insight into how, within the bounds of our ethical standards, BAT can give support to politicians which are key to the final decision-making (sponsorships, trips to the UK, interaction with Foreign Office and Westminster).” 582

Marketing plans The most fascinating insights into BAT’s strategy come from its marketing plan.423 BAT wanted to increase domestic market share from the 75% then held by the state monopoly to over 80% within 4 years, by ensuring a manufacturing monopoly and then closing the market to external competition:

“Our aim is to conclude agreements with the Moldovan government such that fiscal policy can be shaped to deter competition from importing cigarettes in large quantities, thus protecting the domestic cigarette business and ensuring excise revenues to the government from this source. We expect cooperation from the relevant authorities will restrict imports to no more than 20% of the market.”

This approach was elaborated further:

“We must confirm status as monopoly producer of cigarettes, It is imperative that we obtain guarantees from the Moldovan government on domestic market protection by enforcement of fiscal measures using Banderoles” 423 BAT’s excise and pricing strategy outlines how it aimed to achieve this by ensuring import tax exemptions for BAT brands. Though hoping for a monopoly position, BAT

247

assumed it would have freedom to price cigarettes (a situation it achieved in Uzbekistan despite monopoly status there – see Chapters 9 & 11),491 although it knew this might be blocked by the anti-monopolies committee. Through lobbying, BAT also sought an excise system favouring its brand portfolio, to encourage smokers to transfer from filterless to more expensive filter brands, despite their low disposable incomes.423

This approach was to be supported by focusing advertising. Although brand strategies would be developed to capture all segments of the market, a greater focus was to be placed on international filter brands with less marketing support for cheaper brands. For the most expensive international filter brands the aim was to specifically target young people, with tactics such as:

• “A highly selective distribution policy in early years to reach “opinion leaders” Y.A.U.S. [Young Adult Urban Smokers]

• “Highly targeted communications support programme aimed at Y.A.U.S.”423

Despite plans to establish a monopoly, BAT’s communications strategy noted that:

“Investment levels will be made as if operating in a free market with no “monopoly” guarantees extant as there remains the need to build brands in advance of possible advertising restrictions, cessation of monopoly position and the continuing existence of competitor imports at aggressive prices.”423

BAT developed strategies to prevent or undermine effective tobacco regulation:

“It is our intention to spend on advertising while media is available, legal, relatively cheap and uncluttered. Concerning the future, the situation regarding the advertising environment is unclear in terms of restrictions, which may affect the mix of communications spend.

Given the proximity to European media, it is reasonable to assume increasing influence of anti-smoking lobby and associated advertising restrictions. Therefore, as part of our strategy we shall work with the relevant authorities in developing voluntary codes, lobbying activities in order to pre-empt imposition of draconian measures.”423

248

On this basis, BAT estimated that the market would grow from an estimated 5.6 billion in 1994 to approximately 7.5 billion in 2003 and per capita consumption from 1,250 to 1,650 cigarettes per year. BAT assumed that market growth would occur through increased incomes, improved distribution and most tellingly “higher incidence” (i.e. new smokers), particularly amongst women.423

Leaf processing and employment Few details are available on plans for the leaf business but they suggest that through restructuring, the processing workforce would be culled from 2,220 in 1993 to 1,000 by 2003.419 Although other documents suggest that overall jobs would increase, most of the increase would be in seasonal agricultural jobs and the loss in permanent industrial jobs.588 Unsurprisingly this was a cause for concern in Moldova588 and may have been the reason why BAT later suggested that such information be withheld from the Moldovans:

“[a] “sanitised” version of the Leaf Production Programme and supporting assumptions/scenarios will be used in presentations to the Moldovans to avoid protracted negotiations on future levels of employment and the like.”581

Although I was unable to obtain “sanitised” versions of the Leaf Production Programme to assess to what extent this plan was implemented, another document relating to the leaf processing programme refers to a “Moldovan Version” 594 suggesting that alternative versions were indeed produced.

An update Although the documents described above date largely from the mid-1990s, BAT’s recent activities suggest it maintains an active interest in the Moldovan market. In March 2001, on the eve of World No Tobacco Day, BAT launched one of its so called “youth smoking prevention” campaigns, designed to convey the message that smoking is an adult, and thus desirable pursuit, attempting to dupe government ministers to back it and paying TV channels to cover its press launch. However, in the face of concerted opposition, the campaign was abandoned.595 BAT’s efforts rapidly resumed however. Recent examples include a 2003 lottery campaign “let’s win with Pall Mall”, for which the prize draw (for two new Renaults, of great value in Europe’s poorest country) once

249

again coincided with World No Tobacco Day and sponsorship of a major bowling tournament. 596 However, it is reported that Seita has recently agreed a licensing arrangement with Tutun CTC and may now also have an interest in the market.597

Reemtsma/Imperial currently has no direct presence in Moldova.598 Moreover, recent reports suggest that in response to pressure from the UK government to reduce smuggling, Imperial has reduced exports to Moldova and other high-risk countries, indicating that Moldova may have been used as a smuggling post for Imperial products.599

DISCUSSION In the debate about tobacco industry privatisation, this chapter highlights the problems of political interference that beset state owned industries, compounded by the problems of corruption in countries like Moldova. Simultaneously it highlights the dangers of privatisation particularly when investing companies are willing to collude in this corruption, to use anti-competitive business practices that fall short of recommended standards549 and to market their products with the express aim of increasing consumption.

These documents suggest that both Reemtsma and BAT sought to avoid a competitive tendering process which, as the latter noted elsewhere, would be likely to increase the price.480 In so doing they would have reduced the main benefit of privatisation. BAT wished to establish a manufacturing monopoly in Moldova and to close the market to external competition by persuading the government to implement tariff barriers. It planned to lobby the government to influence the excise system to its advantage and prevent the imposition of advertising restrictions, encouraging instead a voluntary code. High levels of marketing were planned despite BAT’s hopes to establish a monopoly position, and advertising was to particularly target young people in cities. BAT clearly thought these strategies would increase the incidence of smoking, particularly among women, directly belying the industry claim that advertising only encourages brand switching.

250

The main limitation of this chapter is that, as privatisation did not proceed, it remains impossible to prove that BAT would have carried out the plans it outlined. However, previous work based on internal tobacco industry documents suggests the industry is usually successful in implementing such plans. Voluntary codes were implemented in Russia and Ukraine (Chapter 2) and smoking surveys indicate that smoking prevalence has increased particularly amongst young women in cities, in line with BAT predictions (see Chapter 4).129

The documents also indicate that BAT, aware of sensitivities around employment levels, would only supply the government with partial truths, intending to hide, for example, the intention to cut jobs in the leaf industry by over half. Furthermore, evidence from Uzbekistan suggests that working conditions of those still employed is likely to deteriorate from what is already a precarious situation.600

It is evident from the tactics revealed here that countries undertaking tobacco industry privatisation and the organisations advising them need to ensure a transparent process and a truly competitive tender in order to maximise potential gains. To minimise the harms they should assess joint venture and privatisation proposals and their true impacts on employment more cautiously, and seek to prevent the predicted increase in consumption likely to arise through the growth of advertising and decline in prices by implementing effective tobacco control policies particularly comprehensive advertising bans and adequate taxation rates. The case of Thailand, which was forced to open its market to cigarette imports as a result of a GATT ruling, shows that comprehensive advertising bans and steep tax increases can be implemented despite opposition and help control increases in smoking prevalence.103 Cigarette consumption, however, continued to increase until the economic crash of 1997,601 not least because of the TTCs aggressive tactics, slashing prices and exploiting legislative loopholes.103

The findings also raise a more general issue about the role that international financial organisations play in pushing for tobacco industry privatisation. I would argue that empirical studies of the health and economic impacts of privatisation are needed to properly inform this debate. These should inter alia examine the impacts on employment, trade balance (given the TTCs use of non-local leaf), demand creation and

251

the long term impacts on health and economy. Meanwhile a precautionary approach should be pursued. Ideally, health impact assessments should be undertaken before individual privatisations. Loan conditions should ensure that public health is protected and corrupt TTC and government activity minimised. Such objectives would be advanced by making the implementation of tobacco control policies and open, competitive tenders pre-requisites for privatisation. Otherwise these organisations will simply serve to propagate the TTC’s relentless expansion and exploitation of yet more vulnerable populations and the further spread of the global tobacco epidemic.

252

CHAPTER 9 The invisible hand: how british american tobacco precluded competition in Uzbekistan

BACKGROUND

Uzbekistan and Central Asia Uzbekistan is the most populous of the five central Asian republics which, with Kazakhstan, Kyrgyzstan, Tajikistan, and Turkmenistan have a combined population of approximately 55 million. These states differ in many ways from the ten other countries to emerge from the former Soviet Union having, for example, younger population structures, different mortality patterns and generally weaker economies.43,23

Most of the ex-Soviet states undertook rapid and large scale privatisation. Uzbekistan generally adopted a more cautious approach, leading it to flounder towards the bottom of the World Bank liberalisation index36 along with Tajikistan (hindered by civil war post- independence) and Turkmenistan (with Uzbekistan, the least democratic of the central Asian states).25

BAT’s 1994 acquisition of the previously state owned tobacco monopoly in Uzbekistan was therefore one of the country’s few privatisations, and the largest to date, 602 accounting for over 30% of all foreign direct investment into Uzbekistan between 1992 and the end of 2000 (Chapter 2). Transnational tobacco companies (TTCs) also made major investments in neighbouring Kazakhstan and Kyrgyzstan where, certainly in the former, tobacco industry privatisation was, in contrast, one element of a wider reform process.35

Since independence, President Islam Karimov has held almost absolute power in Uzbekistan, his autocratic rule increasingly criticised for its poor human rights record and systematic use of torture603,604,605 most recently involving the shooting of a large number of demonstrators in the city of Andijan. 606 Despite largely rejecting international advice to pursue rapid and extensive privatisation, Karimov aligned himself closely with the BAT deal, then Central Asia’s largest foreign investment,221,222, 607 hoping to use it to project Uzbekistan as a safe environment for investment.608

253

The Transition debate and corporate conduct As outlined in Chapter 1, a policy of rapid and extensive privatisation of state owned enterprises was central to the Washington consensus promoted by the multilateral financial organisations and the US Treasury, and became a key element of the economic transition throughout the former Soviet Union. Its proponents assumed that rapid privatisation would be an effective driver of reform and regulatory structures could emerge later.

The largely disastrous consequences of such rapid reform for much of the former Soviet Union (FSU)30,609 have precipitated a number of analyses of the reasons for failure as outlined in Chpater 1, Section 1.4.34,37 These suggest that the environment within which privatisation takes place, including macroeconomic stability, hard budget constraints, competitive markets and adequate property rights, is crucial.29 In the FSU, where such conditions were overwhelmingly absent, privatisation brought few benefits.37

Such analyses have, however, largely focused on system and infrastructural failures including the role of weak or corrupt governments, while the role that transnational corporations (TNCs) may have played has generally been overlooked. Instead, the literature largely sees TNCs as victims of, rather than potential contributors to, the problems. Given that TTCs were among the first and largest investors in the FSU, the release of internal tobacco industry documents through litigation362,360 provides a unique opportunity to address this research gap by exploring the influence of TNCs on the privatisation process.

Chapters 6-8 have explored BAT’s general approach to the opening of the FSU markets, raising concerns, inter alia, about its desire to establish monopoly positions and its anti- competitive approach. This chapter looks specifically at BAT’s corporate behaviour in Uzbekistan, exploring its influence on the privatisation process and how it achieved its monopoly position (subsequent chapters explore BAT’s influence on policy in Uzbekistan). In particular this chapter examines the extent to which BAT’s behaviour could be considered anti-competitive. According to the World Trade Organization, approximately 80 of its member countries, including some 50 developing and transition countries, have adopted competition laws, also known as “antitrust” or “anti-monopoly”

254

laws.610 These deal with a range of anti-competitive practices – practices which restrict or eliminate competition in a market, particularly if employed by a dominant firm - and include such things as absorption of competitors, exclusive dealing and erection of barriers to market entry.610, 611 , 612 Although Uzbekistan’s anti-competition laws are poorly developed,613 OECD guidelines for multinational enterprises working from or in OECD member states (and thus including BAT) outline standards on competition which such enterprises are expected to follow.549 Though such guidelines and the business standards to which the BAT itself now claims to adhere (Table 9-1) were developed after the events discussed here, they nonetheless provide useful benchmarks against which to assess BAT’s conduct in Uzbekistan.614 Examination of BAT’s behaviour also enables an assessment of the extent to which its investment helped address macroeconomic problems, transform economies, and promote efficiency and growth, as the IFOs intended when promoting privatisation.

This work assumes particular consequence since issues of corporate conduct are increasingly important to TTCs. They form the basis of any claim to ongoing legitimacy as stakeholders given both the scale of tobacco’s health impacts and the World Bank’s demonstration of the economic benefits of tobacco control327 and underpin the TTCs efforts to reinvent themselves as responsible companies through their burgeoning corporate social responsibility initiatives.561,615

255

Table 9-1 Business practice standards on competition. On the left - BAT’s business conduct standards on competition law (source: reference 614 page 6). On the right – OECD standards (source: reference 549). Emphasis (underlined) added BAT’s business conduct standards OECD standards Group companies will ensure that they comply with IX COMPETITION the competition laws of each country and economic Text area in which they operate. This is not only because Enterprises should, within the framework of applicable compliance is required by law but also because British laws and regulations, conduct their American Tobacco believes in free competition. activities in a competitive manner. In particular, enterprises should: It is the responsibility of directors and managers of 1. Refrain from entering into or carrying out anti- Group companies to be aware of and familiarise competitive agreements among competitors: themselves with any competition laws affecting their a) To fix prices; companies and their markets and to ensure b] To make rigged bids (collusive tenders); compliance within their organisation. c) To establish output restrictions or quotas; or d) To share or divide markets by allocating customers, Competition laws are intended to promote a free and suppliers, territories or lines of competitive market-place and it is in the interests of all commerce; participants that they are complied with. 2. Conduct all of their activities in a manner consistent with all applicable competition laws, taking into account The competition laws of most countries affect both the applicability of the competition laws of jurisdictions ‘horizontal’ agreements, that is, those between whose economies would be likely to be harmed by anti- competitors, and also ‘vertical’ arrangements between a competitive activity on their part. supplier and its customers. Horizontal price-fixing 3. Co-operate with the competition authorities of such agreements among competitors are likely to be jurisdictions by, among other things and subject to considered amongst the most serious offences, with very applicable law and appropriate safeguards, providing as heavy penalties for infringement for the company and prompt and complete responses as practicable to possibly for the individual involved. In the UK, for requests for information. example, imprisonment can be imposed, as well as 4. Promote employee awareness of the importance of heavy fines. compliance with all applicable competition laws and policies. Most competition laws are likely to impact on joint ventures and all prohibit abuses of dominant position. 55. ……… The term “competition” law is used to Many countries also impose merger control, often with refer to laws, including both “antitrust” and a need to notify a proposed merger for approval “antimonopoly” laws, that prohibit collective or before implementation. unilateral action to (a) abuse market power or dominance, (b) acquire market power or dominance by Although the law may be stated simply, the factual means other than efficient performance, or (c) engage in circumstances to which the law must be applied are anti-competitive agreements. sometimes less clear. If, therefore, there is any doubt whether a particular business practice or activity might be 56. In general, competition laws and policies in breach of competition law, the matter must be referred prohibit (a) hard core cartels; (b) other agreements that to the relevant legal counsel. are deemed to be anti-competitive; (c) conduct that exploits or extends market dominance or market power; and (d) anti-competitive mergers and acquisitions……

256

METHODS The detailed methodology is set out in Chapter 5. Here I include only those details unique to this and the following two chapters on Uzbekistan. As part of the broader search for documents on the FSU, over 35 specific terms were included for Uzbekistan. An iterative approach was used with initial broad search terms such as “Central Asia”, “CAR” (central Asian Republics), “Uzbek*”, “Tashkent”, later narrowed to include the names of key individuals, places, projects, factories and so on identified.

302 documents were coded as relevant to Uzbekistan and indexed fully in the project database. The documents used in this chapter were then extracted from the database by retrieving documents relating to Uzbekistan in chronological order to construct a timeline of events.

To contextualise and triangulate findings, documents on relevant Uzbek legislation were subsequently identified via the on-line BAT Documents Archive established in 2004,362 additional information was sought from tobacco industry journals, the United States Department of Agriculture, BAT Uzbekistan’s records deposited at Companies House and the World Wide Web. Obtaining in-country information was seriously limited by the political situation in Uzbekistan which deteriorated during the course of my work and precluded the possibility of a research visit. Substantial efforts to contact a variety of individuals including Uzbek and non-Uzbek citizens who may have had knowledge of the situation or access to related legislation proved difficult although a small number of individuals closely involved in events in the 1990s were interviewed by telephone and served to confirm my findings. For their own safety however, these individuals cannot be identified here. It is notable in this respect that those who have previously been active in bringing tobacco control issues to the media’s attention in Uzbekistan have been subject to harassment and have left the country for their own safety. I also contacted those who were active in international tobacco control in the mid-1990s to ascertain external knowledge of or opinion on events there.

257

RESULTS The Uzbek Tobacco Industry At independence, the Uzbek tobacco industry consisted of a single tobacco factory – Tashkent Tobacco Factory (TTF) and two fermentation plants in Urgut and Samarkand (UFP and SFP). The market was severely undersupplied: in 1993, TTF was only producing 3-4 billion of an estimated total demand of 22 billion cigarettes.321,616

Key players in the Uzbek tobacco industry’s privatisation were UzPisheProm (UPP), the Uzbek Food Industry Association which was responsible for the tobacco sector, and the State Privatisation Agency (GKI) which owned all state assets.617 UPP staff included its Director, Mr Khamidov and Deputy Director, Mr Husnutdin Usmanov.486 The Cabinet of Ministers also played a key role as a Cabinet decree was necessary to permit any large scale joint venture.617 A timeline of the events described in this chapter is set out in Table 9-2.

Initial Approach: committing the Uzbek authorities to negotiating exclusively with BAT BAT’s first contact with the key players in Uzbekistan came in April 1993 when Ton van Waay (Senior Manager, New Business Development) and Guy Harrington of Schroders (acting as BAT’s advisers) were invited to join a World Economic Forum visit to central Asia. This visit, which included meetings with key officials from heads of state downwards,618 was to prove pivotal, enabling van Waay to arrange a team visit a few weeks later.619

BAT’s initial plan, approved by the Chief Executive’s Committee (CEC),620 was to contract TTF to manufacture cigarettes using BAT-owned machinery, and thereby to:

“lock the Uzbek authorities into an agreement to negotiate a JV [joint venture] exclusively with BAT by the simple expedient of demonstrating quickly a genuine commitment to assist the Tashkent factory. The proposal,…would not involve BAT in major capital expenditure, would allow us to gain first hand experience of trading in Uzbekistan whilst developing our brands in a significant new market, and would provide BAT with an exclusive platform to negotiate a JV in Uzbekistan ahead of competition.”621

258

This proposal was rejected by the Uzbeks who instead offered BAT 49% ownership of TTF in return for the machinery and one year’s exclusive negotiations for a joint venture with the entire Uzbek industry.622 Exclusivity would commence on signing a Letter of Intent,622,623 otherwise known as a Memorandum of Understanding (MOU),624 on which BAT immediately set to work. 403,617,625,626,627,628,629,630,631,632 As described below in more detail, the MOU was signed a few months later and included a 12 month exclusivity period precluding the Uzbek party from initiating discussions with any other party.633 BAT successfully used this clause to pressure the Uzbeks not to engage in negotiations with potential competitors,496,634,635 repeatedly stressing its importance.636,637

259

Table 9-2 Time line of events

April 1993 BAT participates in World Economic Forum visit to Uzbekistan May 1993 First BAT team visit to Uzbekistan. June to October 1993 Ongoing visits and negotiations. September 1993 Philip Morris’ joint venture with Almaty Tobacco Kombinat announced 28th September – 5th October 1993 Uzbek delegation visit UK and Europe. 5th October 1993 Memorandum of Understanding signed between UPP, GKI and BAT. Includes an exclusive negotiating period of 12 months. 23rd – 26th November 1993 President Karimov’s visit to UK. 14 – 16th December 1993 Sir Patrick Sheehy (BAT Chairman) and Ulrich Herter (Managing Director Tobacco BAT Industries) visit Uzbekistan. 15th December 1993 Protocol of Intent signed by First Deputy Prime Minister Ismael Djurabekov for the Uzbek government and Sir Patrick Sheehy for BAT. January 1994 Establishment of customs union between Kazakhstan, Kyrgyzstan and Uzbekistan announced. 11th February 1994 First Presidential Decree: Cabinet of Ministers Decree signed by President Karimov approving joint venture with BAT. Agreement on terms of joint venture to be agreed by 11th May. 14th May 1994 Share Purchase Agreement Signed. 16th May 1994 BAT publicly announces the deal. 20th June 1994 Second Presidential Decree: President Karimov signs Cabinet of Ministers Decree on establishment of joint venture. Confirms monopoly position. June, July, August 1994 Ongoing negotiations focus on taxation 15th July 1994 BAT discovers existence of Health Decree 30, comprehensive tobacco control legislation. Late August 1994 Tax issues resolved to BAT’s satisfaction and a Tashkent decree banning street advertising is reversed at BAT’s request. August – October1994 Negotiations on Health Decree 30 continue. 31st October 1994 Date by which amended decree would be in force 22nd November 1994 BAT transfers first payment to establish majority stake in the Uzbek tobacco industry. 20th – 22nd December 1994 Sheehy visits Uzbekistan to formalise creation of the joint venture

Desire to avoid a competitive tender BAT clearly aimed to avoid a competitive tender, and William Wells of Schroders emphasised in an early briefing note that “[s]peed will be of the essence if the Uzbeks’ confidence is to be secured and the risk of a competitive tender is to be minimised.”617 Competition to acquire assets in Uzbekistan was intense - by the time of BAT’s first team visit in May 1993, the TTF director had already signed 14 letters of intent for joint

260

ventures.486 By September 1993, two related factors served to increase the possibility of a competitive tender. The first was that the Uzbek authorities, having previously been clear about their intent to establish a joint venture,486,628 came to favour outright privatisation.625,627 BAT judged that this reflected the influence of Price Waterhouse (PW), appointed to advise the GKI on privatisation (and, notably, BAT Uzbekistan’s auditors from the company’s inception onwards638):

“the presence of PW as the GKI’s advisers on privatisation strategies and practices may mean that the GKI is being educated - albeit informally - about the merits and demerits of tender sales.”627

The second was the success of Philip Morris International (PMI) in the competitive tender to acquire the Almaty Tobacco Factory or Kombinat (ATK) in neighbouring Kazakhstan. BAT’s main concern was that a tender would “almost inevitably increase the cost of investment”.480 Aware that it had already lost tenders in Lithuania and Kazakhstan,490 (see Chapter 7), BAT described the matter as one of “extreme urgency”625:

“Now that Kazakhstan’s tobacco industry has been awarded to a competitor there is a clear danger that BAT will be forced to compete in tenders at very high prices or shut out of remaining business opportunities in Central Asia. …… BAT was outbid in Kazakhstan by a factor of more than three on payment to the Government for shares of ATK and by a factor of two for capital expenditures. …… If the Kazakhs announce these figures publicly …… the stakes will rise immensely in the remaining republics of Central Asia and perhaps elsewhere in the CIS [Commonwealth of Independent States] as well. ……… … We believe it is essential that good progress toward an agreement with the Uzbeks be made while they are in London. Further delay will increase greatly the risk of a competitive tender in Uzbekistan.”625

Philip Morris’ acquisition of ATK also fuelled BAT’s desire for rapid success in Uzbekistan, which then became “the only remaining large business opportunity in Central Asia”, implying that “[i]f BAT does not establish itself in a dominant position in Uzbekistan it will not be able to compete effectively in Central Asia as a whole.”625 The regional

261

picture would also be influenced by the customs union linking Kazakhstan, Uzbekistan and Kyrgyzstan,480 announced in January 1994.496 BAT’s marketing expert Dean Sims recognised that the company had made a strategic error, having viewed “the Uzbekistan opportunity erroneously as an independent investment” while PMI’s substantially higher offer in Kazakhstan indicated that its investment there was seen “as crucial to the whole region and giving them the possibility to dominate the region and move first in terms of marketing initiatives.”497

Ulrich Herter (Managing Director Tobacco, BAT Industries), feared that there was “a serious risk that unless BAT takes immediate pre-emptive steps to progress negotiations with the Uzbeks the Uzbek tobacco industry will go to competitive tender”.624 His recommendation, approved by the CEC, was “that BAT should aggressively pursue the existing investment opportunity” and sign a MOU as a matter of urgency.624,639 This was ultimately achieved on 5th October 1993,633 the last day of an Uzbek delegation visit to Europe636 and, as BAT had desired,625 focused on the creation of a joint venture.633

Use of political contacts to ward off competitors and prevent a competitive tender Further efforts to secure BAT’s position and prevent a competitive tender relied on its extensive political contacts, and particularly the support of President Karimov. Documents suggest that Usmanov assisted BAT in attempting to move the project forward from the inside, keeping BAT in the forefront and warding off pressure from competitors:437,487

“Government offices are deluged with letters from various consultants fronting for PMI, RJR and Rothmans… I can ward off this pressure, but only so much, - Usmanov said. - Sometimes these letters come to me for expert opinion and I try to keep BAT in the forefront. But I am sure that a lot of them stay in other offices and who knows what other Government officials might think. PMI for instance offers 200-300 mln Dollars and credits and KPMG are lobbying very hard for them. RJR signed with us a Protocol of Intent without an exclusivity clause but with a promise to invest

262

……….These letters go to practically all the important Government addresses complaining about the exclusivity with BAT.”487 By January 1994 Usmanov claimed to have succeeded in securing the exclusion of the tobacco industry from the privatisation programme “for the time being” and advised BAT not to await privatisation since it “would then be dealing with ‘closed’ joint stock companies which he thought would be more difficult.”637

The nature of Uzbek politics meant that presidential approval for any joint venture plans would be critical. Contact with the president did not occur until President Karimov’s visit to the UK in November 1993, 634,640,641,642,643 followed by chairman Sir Patrick Sheehy’s visit to Uzbekistan in December. During this visit, the Protocol of Intent was signed, 480,508,644,645 a crucial document for BAT:

“Usmanov believes the meeting with the President and the Protocol of Intent make life much harder for the competition. Now we can cay (sic) to all the others: go to the President and secure his support, then we will deal with you. So far he is with BAT.”487 At this point pressure for a competitive tender was building both externally from BAT’s competitors, and internally from GKI and its newly appointed advisers on the tobacco industry deal, KPMG (who had acted as advisers to the Kazakhs on the Almaty deal437,496 and who, two documents suggest were also representing Philip Morris in Uzbekistan).487,496 The fact that BAT now had “the active support of President Karimov”,480 who was willing to issue a presidential decree authorising the establishment of a joint venture with BAT,480 was vital to BAT’s anti-competitive strategy. This was evident in a note to the CEC describing Karimov’s response to arguments advanced by GKI that a competitive tender would yield higher investment and greater employment protection:

“Rather than being concerned about the valuation issue, when the President heard of this debate and the possible delays in implementing the investment, he instructed the Deputy Prime Minister to prepare a Presidential Decree authorising the establishing of a joint venture with BAT to be submitted to the Cabinet of Ministers on 1st February for his signature on 4th February.646

263

A presidential decree signed on 11th February 1994496,647 approved the establishment of a joint venture with BAT, based on TTF, SFP, UFP, and including the construction of new cigarette factory and leaf processing plant in Samarkand region. As Herter reported to the CEC, it also confirmed BAT’s sole negotiator status:

“Although due to the Decision, GKI has to accept BAT as sole negotiator at least until 11th May 1994 it will probably look for opportunities to state its case for an open tender if BAT’s proposal is short of its expectations. ……Although KPMG lobbied strongly against BAT having a sole negotiating position they will have to work within the aims and terms of the Cabinet of Ministers Decision.”496 The decree specified that the joint venture terms had to be agreed by 11th May 1994, three months hence.496 This precipitated a rush of activity as BAT began formal negotiations with the Uzbek government and its advisors.648,649,650

Absorption of competitors The inclusion of TTF and both fermentation plants in the joint venture had been a deliberate ploy by BAT to exclude the possibility of domestic competition in what effectively amounts to absorption of competitors. Despite intending to close SFP and ultimately TTF,321,496,616 with UFP alone recognised as “a lynchpin to any investment in Uzbekistan”, 651 BAT took “pre-emptive action” to deny all three to competitors.496,644,651 They achieved this by installing machinery and services at TTF and UFP and including SFP in the joint venture to prevent others acquiring it, noting in relation to TTF:

“[i]n order to provide immediate access to the most concentrated consumer market in Uzbekistan and to deny the market and TTF to BAT’s competitors, it was agreed that it would be desirable to be located in Tashkent for at least the short term.”651

Exclusive dealing and erecting barriers to market entry Having secured sole negotiator status, prevented a competitive bidding process and absorbed potential domestic competitors, BAT then attempted to secure the Uzbek

264

market from competition that could arise through imports or new market entrants. Referring openly to “[p]rotection of the domestic tobacco products market”,641 BAT set about achieving this through exclusive dealing and erecting barriers to market entry.

It requested a number of anti-competitive preconditions to its investment,321,480,496,497,641,652 summarised clearly in the draft Skeleton Business Plan,321 which indicates that BAT aimed to: “secure a dominant position in the market, achieving 80% market share by the year 1997 by concluding a joint-venture with the Uzbek Cigarette Industry which will guarantee our position over the plan period through providing competitive advantage particularly the restriction on imports.”321 BAT sought exclusive cigarette manufacturing rights, 480,496,652 an exclusive arrangement with Bakalea (the state distribution agency),497,641,651 and with local advertising agencies,497 confirmation that the tobacco industry was not subject to legal constraints as a monopoly,641 and reform of the tobacco excise and import tax systems.496,497,622,641,652

Exclusivity of distribution through Bakalea, a one way arrangement that would tie up Bakalea for at least five years but not preclude BAT from using other distributors,321 was seen by Dean Sims as a pre-requisite for any investment. Sims also sought the application of import duties on imported cigarettes, qualified by a 3 year indemnity for BAT imports, calculating that duties of 20-25% would be sufficient “to assure a significant BAT competitive advantage in the sensitive pricing environment” and higher taxes on advertising of imported products.497 He emphasised the scale of the opportunity associated with an investment if these conditions could be obtained:

“It must be absolutely clear that what we wish to buy is not manufacturing assets or brands but an opportunity to dominate the market. The TTF assets and brands are worthless without the above guarantees…… With these guarantees it could emmerge [sic] as one of the most significant and lucrative Group investments in the last twenty years. The impact on Group profits could be considerable. No effort or avenue should be ignored to try to achieve the above guarantees but this must be immediate. This may require unorthodox arrangements to be made

265

with the decision makers but this will have to be weighed against the scope of the opportunity.”497 BAT’s exclusive dealing efforts did not stop with Bakalea. On the marketing front, concerned that it was being outdone,497,653,654,655 BAT wanted to tie up all the key players in advertising to prevent competitor access:

“No advertising agencies in the true sense of the word exist locally. All creativity and printing must be sourced externally. However the limited number of people currently working in this field exercise disproportionate influence and as with financially stable distributors, we should tie them up with exclusive contracts now.”497[original emphasis] Lest these wide-ranging measures be insufficient, to further foreclose the possibility of any competition, BAT also specified that no other domestic or foreign business should be licensed to process leaf (at least until facilities had been installed at UFP) or to manufacture tobacco products (for 5 years or 6 months after the new factory is fully operational)321 and that no local or international brands would be registered without prior consultation with BAT.321

It is also clear that, although BAT planned to establish a monopoly, they hoped to achieve this without being subject to anti-monopoly regulation, as freedom to price was seen as an absolute pre-condition to investment:

“BAT would also require confirmation that the Uzbek Anti-Monopolies committee did not deem that any part of BAT’s investment would be subject to regulation.” and later note that “the Customs Union helps BAT defend itself from accusations of seeking to establish a legal monopoly, albeit short term. It might be appropriate to establish whether or not the terms of the Customs Union permit such exclusivity and what the Anti-Monopoly Committee stance may be.”321

Investment privileges In addition to established foreign investor tax privileges,320,622,641 BAT sought various additional and wide-ranging privileges.641,650,652, 656 It noted that these would lead to considerable savings,656 further bolstering its position whilst reducing potential

266

government revenue from BAT’s investment. For example, despite internally acknowledging that existing legislation provided a 2 to 5 year income tax holiday and that the 5 year exemption being offered was “extremely generous”, 650,657 BAT pushed for an additional five year exemption. Other privileges sought included a five year exemption from taxes on foreign currency income and a ten year exemption from import, customs and excise duties on materials imported for processing.650 BAT also sought clarification that no excise or VAT would be levied against imports that formed part of BAT’s initial or subsequent capital contributions.650 This request appears to stem from the suggestion by Elena Kirillova, of McKennas, that “any future BAT imports could be imported free of import duties provided they were structured as part of a future investment. In order to have such imports considered as capital investment contributions a strong argument that they are required to “build the brand” must be presented.”651

Securing the deal On 14th May 1994, 3 days after the deadline for establishing the joint venture set out in the presidential decree, official agreement was reached in the form of a Share Purchase Agreement 658 ,608 and BAT formally announced the deal.509, 659 On 20th June 1994 Karimov signed a second Presidential decree effectively activating the Share Purchase Agreement and guaranteeing BAT its desired monopoly position.658 Although unilateral changes from a previous agreement were made by the Uzbeks,658,660 the decree gave BAT more or less what it had demanded: exclusive manufacturing and processing rights for 5 years, 458,658, freedom to contract with tobacco farmers, release from the state order for cigarettes and leaf, and receipt of the privileges contemplated in the May Agreement.658 A subsequent USDA report notes that not only were the import duty exemptions granted for five years, they were then extended for a further five, and that BAT was given a 10-year exclusive right to grow, process and export the Turkish leaf variety ‘Izmir’.223 Above all, BAT was given freedom to price its cigarettes whilst avoiding inclusion on the monopolies register, as Nick Brookes, director of new business development noted:

267

“5. We have negotiated a 5 year monopoly for cigarette manufacture in Uzbekistan. This could only be achieved by arguing that competition would be available from imports. 6. Despite being a monopoly we have, nevertheless, negotiated exclusion from the Uzbek Monopolies Committee which amongst other things, would have restricted our freedom to set prices”491 He also admitted that, as a result of BAT’s efforts to redesign the taxation system (described further in Chapter 10), the argument that competition existed from imports was largely spurious:

8. If the new level playing field tax regime is properly applied, cigarettes entering Uzbekistan from outside the CIS should in any event reach the market at a price disadvantage to locally manufactured cigarettes owing to higher ex- factory prices, transportation costs, etc.”491 The fact that the Uzbek monopolies committee deems firms with a market share of over 65% as “dominant”613 while BAT aimed for a share of 80%,616,661 achieving a 72% share by 1999,211 makes BAT’s exclusion all the more remarkable.

Hiccoughs at the final stages Though completion of the deal appeared imminent,662,663 several issues emerged to delay progress. Most notable were those described in the following two chapters, excise reforms and discovery of a tobacco control decree (Decree 30) issued by the Ministry of Health in July.664,665

BAT made considerable efforts to influence levels of import and excise duties, starting before the share purchase agreement was signed and continuing long afterwards, as detailed further in Chapter 10. BAT’s initial plans to have “punitive”641 import tariffs imposed, as alluded to above,497 had been abandoned because the Uzbek government had been highly sensitive to what it saw as anti-competitive practices.491,666 During negotiations it had become apparent “that seeking all three of protective import duties, manufacturing exclusivity and pricing freedom was impractical”, with the former deemed the least necessary during initial stages.666 Instead, BAT sought and ultimately achieved considerable reform of the excise system that bolstered BAT’s monopoly

268

position, reduced competition and favoured its own brands over those of its competitors (see Chapter 10).

In a document that indicates how BAT and Schroders were aware of the dubious nature of such anti-competitive practices, William Wells outlines how he was unable to allay the BAT chairman’s concerns about the failure to pursue punitive import duties during his visit491 because he “was not entirely clear to what extent it was appropriate to talk about what might be construed as anti-competitive practices, in front of Neil Buckley of the Financial Times.”666

A further concern was the potential for controls on advertising and restrictions on smoking in public places. By October, after aggressive and persistent negotiations, ultimately involving President Karimov, BAT had successfully secured exemption from the Health Ministry’s Decree 30 that would have constrained it, 667 , 668 instead implementing its own voluntary code on advertising (see Chapter 11).669

Outcomes By November final agreement on the deal was reached670 with BAT transferring their first payment on 22nd November 1994671 as planned608 to acquire a 51% stake in the Uzbek tobacco industry.670,672 Documents indicate that a total of $60.08 million dollars was transferred to Uzbek accounts at Chase Manhattan Bank,671 the equivalent of £38.2 million at exchange rates at that time, but Companies House records indicate that BAT invested £44.9 million.672 When asked for an explanation BAT attributed the £6.7 million discrepancy to capitalisation of related costs including legal and merchant bankers fees and travel costs incurred in the acquisition (personal correspondence Micheal Prideaux, Director Corporate and Regulatory Affairs July 2005), suggesting, therefore, that such costs contributed a very high 17.5% of the deal price.

Sir Patrick Sheehy and other BAT delegates visited Uzbekistan in December 1994 to formalise the joint venture deal. BAT documents record that they met with President Islam Karimov:

269

“who praised BAT as a solid international partner with a vision, and assured the guests of his continuing personal support for the company’s long-term investment plans ..... According to senior Uzbek officials, Sir Patrick’s trip has greatly contributed to the image of BAT in Uzbekistan as a serious investor and a responsible corporate citizen.”673 Public reports of the deal indicate that BAT made further investments to an estimated total valued at over $300 million (roughly £200 million) by 1998 with its shareholding increasing to 97% (Chapter 2).221,674, Companies House records suggest that of the investments made by the end of 1998, £144 million was cash.638 As planned, its production levels increased gradually until by 1999, export activities had begun675 and BAT’s market share had reached over 70%.211 In 2000 and 2001 BAT Uzbekistan made profits of approximately £1 million and £2.8 million respectively.638 Subsequently however, domestic and export sales have both fallen with declines attributed to greater levels of competition, introduction of import taxes in neighbouring countries and smuggled cigarettes676,677,678 and BAT Uzbekistan has recorded a loss.679,680

DISCUSSION This chapter clearly demonstrates that BAT wielded powerful influence over the privatisation process in Uzbekistan, particularly via the support of President Karimov. It prevented a competitive tender, despite considerable pressure from both internal and external agencies, and established a monopoly, yet used spurious arguments to ensure exclusion from the Monopolies Committee and freedom to set prices. BAT engaged in a broad range of anticompetitive practices that cemented its dominant position and precluded the possibility of any effective competition from either inside or outside the country. Simultaneously, it negotiated incredibly favourable investment terms that resulted in large amounts of foregone revenue for the Uzbek government.

The purported benefits of competition underpin the global drive for trade and investment liberalisation and anti-competitive behaviour is broadly condemned within the global corporate sector. Even BAT’s business conduct standards state “British American

270

Tobacco believes in free competition”.614 Yet in Uzbekistan BAT clearly engaged in a wide range of anti-competitive behaviours including absorption of competitors and exclusive dealing. While its efforts to implement tariff barriers in the form of import duties failed, BAT managed to ensure that the excise system was reformed to benefit the domestic producer and its brands. Crucially, BAT also ensured exclusive rights to manufacture tobacco products and process leaf and ensured that no cigarette brands could be registered without consultation with BAT.

BAT’s conduct in Uzbekistan clearly contradicts the spirit of its self-proclaimed business standards and the OECD guidelines which outline how competition laws prohibit action to “abuse market power or dominance” or “acquire market power or dominance by means other than efficient performance”(Table 9-1).549 BAT’s conduct also stands in marked contrast to TNC’s usual calls for the removal of tariff and non- tariff barriers to trade. Indeed in the 1980s TTCs used such arguments to persuade the US government to pressure Japan, Taiwan, South Korea and Thailand to open their markets to imports alleging that import quotas, high taxes and other restrictions unfairly limited the market to US tobacco products (Chapter 1, Section 2.4).92,98

Despite the TTC’s public position on competition, and their willingness to use arguments for competition when they suit, their indulgence in anti-competitive behaviour is not unique to Uzbekistan. Chapters 7 and 8 have shown how BAT successfully avoided a competitive tender in Ukraine, and attempted to do so in Russia and Moldova, in the latter again hoping to shore up its position with protective excise policies. Such conduct should also be viewed alongside evidence from the company’s documents of collusion to agree prices of brands with ostensible competitors. One report indicates that TTCs colluded to fix prices in as many as 23 countries across Africa, Asia, Latin America, Europe and the Middle East.681

Importantly, this thesis also sheds light on whether investment liberalisation and privatisation helped address Uzbekistan’s economic problems as the multilateral financial organisations had intended. It suggests that Uzbekistan, despite its dire economic position, failed to capitalise on the sale of its tobacco industry, seemingly being cajoled by BAT into foregoing a competitive tender and providing BAT with

271

numerous tax holidays and favourable excise policies that substantially reduced revenues accruing to the government. These diverse exemptions and privileges secured by BAT appear to conflict with even the limited requirements of Uzbekistan’s law on foreign investment as amended in July 1992 which identifies broad obligations to “comply with the law applicable in the territory of the Republic of Uzbekistan” and to “pay taxes, make foreign-currency allocations and other payments in accordance with the law of the Republic”.682 Once again, such exemptions are not unique to Uzbekistan, the TTCs also secured 5-10 year exemptions from profit tax in Ukraine,556 Kyrgyzstan and Hungary.683

BAT’s behaviour in influencing the privatisation process, acting anti-competitively and securing investment privileges was undoubtedly facilitated by the absence of effective governance in Uzbekistan, highlighting the importance of building the necessary infrastructure prior to privatisation. The documents outline how BAT exploited the absence of a competitive framework and exacerbated the problems of poor governance. The Uzbek government was clearly complicit in the process, but BAT appears the more culpable given its systematic exploitation of the naivety and inexperience of those with whom it dealt, the poor understanding of competition and regulation,613 and limited regulatory capacity of the newly independent state. Certainly the Uzbeks appeared to have little knowledge of due process, only picking up small details once Price Waterhouse was appointed.

These findings in some respects complement those of Hellman and others who, working from the basis that poor standards of governance and high levels of corruption in transition economies have discouraged investment, analysed the behaviour of firms that had chosen to invest.684 They found that foreign firms investing in the region were significantly more likely than local firms to engage in certain forms of corrupt behaviour and enjoyed substantial benefits from doing so. In other words, such firms contributed to problems of governance through their own poor standards of conduct.

It could, of course, be argued that BAT’s behaviour is acceptable practice in pursuit of the objective of maximising returns to its shareholders. However, this narrow interpretation of corporate behaviour, in which the ends justified the means, is no longer

272

viewed as acceptable, even by BAT, as set out in its own public statements on business conduct614 (Table 9-1) and is at odds with BAT’s claims of corporate social responsibility in Uzbekistan and elsewhere.685

The divergence between BAT’s public statements and its private misbehaviour put those that support the company in an invidious position. The World Economic Forum claims to be “committed to improving the state of the world”686,687 yet its role in launching BAT’s entry to Uzbekistan, is surely antithetical to this ethos. The British Government played a role in promoting BAT in Uzbekistan688 which, as a signatory to the OECD standards, it may wish to reflect on. But perhaps more serious is the role of Schroders and its staff who, in acting as BAT’s advisers, were clearly knowledgeable of and, it seems, active in supporting the behaviours outlined in this chapter as well as those outlined in the following two chapters. It is also noteworthy that two major accounting firms, Price Waterhouse (now Price Waterhouse Cooper) and KPMG appeared to have potentially conflicting roles in the processes outlined. PW was appointed to advise GKI on privatisation, yet a few months later was acting as BAT Uzbekistan’s auditors. PW also took over from Coopers and Lybrand as auditors for the parent company, BAT Industries’ from 1998 onwards, acting in the same capacity for the newly created British American Tobacco plc from its inception. KPMG was appointed by GKI to advise on the tobacco deal whilst also, documents suggest, lobbying for Philip Morris in Uzbekistan. KPMG was unable to confirm or refute this information, being unable to trace any record of KPMG in the UK working on tobacco privatisation in Uzbekistan. Moreover, despite being given copies of a document which stated that KPMG were “lobbying very hard” for PMI, claimed “it is not uncommon for ‘Big Four’ firms to provide due diligence to potential purchasers where the firm is acting as advisers to the (vendor) government.” (personal correspondence Judith Dow, KPMG London,

September 2005).

In summary, these findings may add weight to those who oppose tobacco industry privatisation.10,39 Not only does privatisation appear to encourage rising tobacco consumption with inevitable consequences for health and subsequent indirect negative economic impacts, (Chapter 3),108 but its purported economic benefits may also not be realised. More broadly, these findings also lend support to those who argued for a

273

gradual approach to privatisation, showing the ease with which a leading TNC was able to mislead the Uzbek government and thus act in an anticompetitive manner, highlighting that such behaviour was detrimental to that country’s economy and systems of governance. Had institutional mechanisms to ensure competition and good corporate governance been in place, it would have been far harder for BAT to have behaved in this way. Whilst not refuting the view that weak and corrupt state performance is a key factor explaining the failure of privatisation in the FSU, our findings also indicate that the behaviour of TNCs is also an issue. They also raise further questions about the conduct of a major international company and by implication, those who support it. Although BAT’s disregard for the welfare of those who consume its products is widely recognised, its business practices may be less well known. The inadequacy of self- directed corporate social responsibility initiatives is effectively illustrated by the failure of BAT Uzbekistan’s social report685 to make any acknowledgement of these extremely dubious foundations on which its local dominance has been established and to instead suggest that BAT will seek still more favourable conditions by referring to the need for “more active work for the purpose of providing the Company equal conditions with major competitors and forestall their plans.”685 It is clear that TNCs and perhaps particularly TTCs must be more closely regulated through enforceable codes of conduct. This and the conduct of transparent tenders would go some way towards addressing the problems this chapter identifies.

274

CHAPTER 10 "[A]ssisting Governments to design and implement the most suitable indirect tax systems”: BAT and tobacco taxation in Uzbekistan

INTRODUCTION Increasing the price of tobacco products has been demonstrated to be the single most effective means of reducing their consumption. 689 Evidence of greater price responsiveness among economically disadvantaged690 and young smokers691 further justifies the central role of taxation within health policy. A 10 per cent increase in cigarette prices would reduce consumption by 8 per cent in low- and middle-income countries (LMICs),692 double the projected reduction for high income countries and particularly significant in the context of tobacco’s shifting global burden. 693 Accompanying increases in government revenues would, in the World Bank’s words, bring “unprecedented health benefits without harming economies”.327

From the transnational tobacco companies (TTCs) perspective, increased taxation constitutes a threat to future profits and success. Chapter 7 highlighted the TTCs desire to shape the design of taxation systems in transition markets with BAT establishing a team specifically to advise on excise. Arguably the most egregious example of tobacco industry influence over taxation policy emerges in Uzbekistan where, as outlined in Chapters 2 and 9, British American Tobacco (BAT) established a production monopoly, in a deal announced in May 1994 and finalised in late 1995.

Continuing with the Uzbek example, this chapter aims to explore how investment liberalisation, privatisation and TTC investment may influence tobacco taxation policy. As previously outlined, the International Monetary Fund (IMF) and others have encouraged privatisation of state-owned tobacco companies to help address macroeconomic problems and promote economic growth.10 Yet, as highlighted in Chapter 3, tobacco industry privatisation will tend to lower cigarette prices, thereby encouraging consumption, with serious implications for public health and potential economic and development costs. Taxation provides the means to control or prevent the price fall. Hence, understanding whether privatisation and TTC investment jeopardise effective taxation policies is essential. This is particularly so, given that BAT has used tobacco’s revenue earning potential to encourage LMICs, including Uzbekistan, to accept its investment (Chapter 7).

275

BACKGROUND BAT’s efforts to shape the development of taxation in Uzbekistan occurred within the context of a broader commitment in the 1993-1997 Company Plan to “actively seek to influence governments with regard to the level and structure of tobacco taxation in order to promote market growth and to secure competitive advantage.” 694 A subsequent paper outlining group policy on indirect taxation exhorts the pursuit of “(e)very opportunity to reduce the level of taxation” ,695 with strategic emphasis on the development of contacts with politicians and officials:

Government officials responsible for tobacco excise and VAT planning and control… should be identified and sufficient regular contact maintained while Ministerial (Government and Opposition) contacts should also be maintained to ensure that the Company is well placed to have its views taken into consideration… Such relations should establish BAT as the Company to which Government will turn when they need advice and assistance upon any aspect of excise taxation….”695 Given the geographical diversity of BAT’s operations and the considerable price variation across its brand portfolio, the company did not pursue a globally consistent preference between the possible structures of excise tax (Box 10-1): "There is no single excise structure appropriate to BAT on a global basis because the competitor and brand mix situations vary widely. … … In most instances, long term profit optimisation will normally result from a totally or predominantly specific excise taxation structure, in preference to an ad valorem one. The optimum balance between specific and ad valorem will depend upon the brand mix …. … … In markets where BAT has a dominant share, or where competition is limited because restricted access to the market keeps out imports, our policy is to persuade Government to minimise imports through the application of high tariffs and other obstacles such as pre-payment or early payment of excise taxes…” BAT’s preferred taxation regime for Uzbekistan was to be shaped by a distinctive competitive environment which is examined in more detail in Chapter 9. Only about a fifth of the estimated 14-22 billion cigarettes consumed were domestically manufactured.696,616 While developing its manufacturing base, BAT would therefore

276

have to rely on imports until domestic production increased.616 Smuggled and other cheap cigarette imports were rife and posed a threat to BAT,458,666 as did Philip Morris’ 1993 acquisition of the Almaty Tobacco Factory in neighbouring Kazakhstan, which in January 1994 had established a free trade area with Uzbekistan.697 Unlike BAT, Philip Morris has a far narrower brand portfolio built around Marlboro, a high cost brand.

Box 10-1 Tobacco Tax Structures Ad valorem tax: a percentage of the retail price Tends to widen price differentials by making expensive brands relatively more expensive. It offers governments the advantage that tax is automatically increased with industry price rises. Allows industry the advantage of controlling the tax level by keeping its prices low (industry can lower its prices in response to a tax increase, and prevent any public health benefit). Not generally favoured by the large transnationals who tend to sell the expensive brands. BAT documents indicate that it does not generally favour ad valorem rates, but does so in Uzbekistan to benefit its own low price domestically produced cigarettes and to disadvantage its competitors’ high price imports.

Specific tax: a fixed tax per cigarette Reduces price differentials by adding a fixed tax to every cigarette regardless of its baseline price, thus benefiting more expensive cigarettes and leading to cheaper brands possibly being withdrawn from the market. Offers industry the advantage of raising its base price and profit without the tax increasing, and governments the advantage of being able to substantially raise price and tax revenue by a known amount; therefore generally favoured for tobacco control. Specific taxes are generally also favoured by the large transnationals with expensive brands.

De-minimus tax: a set minimum tax per cigarette A specific tax set if and only if, the ad valorem tax does not reach a set minimum level. BAT advocates for Uzbekistan to ensure that cheap foreign imports, relatively unaffected by the ad valorem element, are rendered less competitive with BAT’s domestic production.

Tax marks/stamps/banderoles These are alternative ways of indicating on the packet that tobacco taxes have been paid. BAT advocated their use in Uzbekistan to ensure competitor’s imported cigarettes are taxed.

METHODS The methodology employed is set out in Chapters 5 and 9. Documents focusing on “pricing and taxation” and “policy influence” in Uzbekistan were retrieved from my database and sorted by date to construct a chronology of events. To ensure that all relevant documents were retrieved, a secondary search of all indexed documents on Uzbekistan was performed. In addition to the efforts outlined in Chapter 9, to contextualise and triangulate findings, documents on taxation in other FSU countries were retrieved from the database and documents detailing broader company policies on taxation were subsequently identified via the on-line BAT Documents Archive

277

established in 2004 using combinations of search terms including “taxation”, “excise”, and “company plan”.

RESULTS BAT’s strategy for its investment in Uzbekistan BAT noted early on that “[t]he financial attractiveness of this proposal is highly sensitive to the tax structure, particularly to the excise regime” 621 and indicated that excise and import duty reform were “[e]ssential elements in any BAT investment”.321 Under the existing system, cigarette imports were actively encouraged to counter inadequate domestic production648- no import license was required698 and no import duties or excise taxes were levied on imports,648,699,700 although the Soviet practice of cigarettes produced in one CIS (Commonwealth of Independent States) country being taxed before being free to move to another continued.699 Excise696 was however applied to domestic products according to cigarette class, with lower quality unfiltered cigarettes enjoying a tax advantage (Table 10-1). 321,651,700,701

BAT’s strategy, which emerged somewhat erratically over time, included four main strands: to introduce import taxes and, ideally, equal excise tax on imports and domestic production;320,666, 702 to develop an excise regime to advantage BAT’s brands;321,702 to reduce excise rates;660 and to ensure the proper control and collection of taxes, particularly on competitors’ imports. 320,513,702

Table 10-1 Changes in ad-valorem cigarette excise rates over time as a result of BAT’s influence Cigarette class and Excise rate (%) at Excise rate (%) post Excise rates (%) brand examples321 outset700,701 government increases proposed by BAT in In March 1994699 June 1994660 and in place by February 1995 696 Class 1 (eg Uzbekistan 40 90 40 – filter brand) Class 2/3 (eg Astra – 25 56 25 plain) Class 4 (eg Risk – 15 34 plain) Papirossy 20 45

Initial focus: protective import duties The company’s initial priority was to secure the imposition of import duties, in BAT’s words - “agreement to protect the domestic tobacco products by use of punitive import duties”,641 although excise reforms were also desired.622,641 The

278

Uzbeks appeared to acquiesce quickly.634 BAT’s Chief Executives Committee (CEC) noted on 29th November 1993 “[t]he Uzbek government … has indicated that it is willing to undertake reforms, including lower excise rates and the introduction of appropriate import duties.”320 Despite this promising start and BAT’s ultimate success, the process was to prove more tortuous than initially anticipated.

Seeking and successfully influencing contacts In December 1993 Sir Patrick Sheehy (BAT Chairman) promised Uzbek President Islam Karimov “a team of excise experts to advise and assist the Uzbek authorities”.508 These experts, David Bishop and Chris Dufty, visited in January 1994644 aiming to make “contact with relevant Government officials to ascertain whether it is realistic to make changes if necessary to indirect tax regulations so as to suit BAT brand marketing strategies”.700 They emphasised potential taxation revenues, a prospect that clearly interested the Uzbeks.513,700 Yet in calculating such revenues, they carefully selected rates that would be unlikely to increase cigarette prices.513

BAT aimed to work closely with the Taxes Ministry702 via key contacts, most notably the Deputy Minister of Tax Inspection, Abdulla Mamasaatovich701 and the Deputy Minister of Finance, Dr Abdoulla M Abdoukadirov,660 seeking to exploit their inexperience and naivety.657,660,700, 703 , 704 Bishop and Dufty noted that Abdoukadirov recognised the importance of levying import duties to raise revenue and protect the local industry: a 50% import duty on cars had already been introduced to accompany joint ventures with Mercedes and Daewoo. Abdoukadirov volunteered similar protection to the cigarette industry and “confirmed his complete willingness to work with BAT on an overhaul of the tax system should any joint venture go ahead.”700 William Wells, of BAT’s advisors’ Schroders, recorded that:

“the excise authorities appear very receptive to excise proposals………. … the impression was that BAT could have almost any exemption it wanted.”637

Emergent obstacles Despite this apparent progress, at a meeting with the Ministry of Finance in February 1994, BAT learned that previous promises of “full cooperation with BAT in order to introduce a suitable indirect tax structure to support and develop domestic manufacture of cigarettes in Uzbekistan” would not be realised.703 Instead, the

279

Uzbek government had introduced a number of changes contrary to BAT’s proposals.703 Most notable was a Presidential decree banning import duties which, driven by the country’s need for imports, had been introduced in February without BAT’s knowledge.491,705 BAT was also disappointed to learn that a 1st March decree would increase cigarette excise rates.657,699 The new rates represented an approximate 2.25 fold rise from previous levels for all four cigarette categories (Table 10-1) with class 1 cigarettes to be taxed at 90%.699

Formal negotiations on a Share Purchase Agreement commenced against this background. BAT ensured that the issue of excise reform was prominent and sought Ministry of Finance representation on the lead negotiation team claiming that “an important part of BAT’s conditions for investment relate to fiscal and excise matters.”650 It is more likely that BAT’s developing relationship with Abdoukadirov underpinned this request, given his apparent willingness to follow its advice657,704:

“Dr Abdoukadirov continued to be open and friendly, was appreciative of the information provided by BAT on international indirect tax treatments and very keen that BAT should work with him and his officials on the drafting of indirect tax legislation and the setting up of collection and control procedures, which he admits they have neither the personnel or expertise to do.”703

A change in approach – away from import duties to equal excise treatment Following the decree banning import duties, BAT and its advisers suddenly changed their approach to ensuring the equal excise treatment of imports and domestic production.491,656,666,699,705 It was felt that this should be sufficient to secure BAT’s competitive advantage:491,666

“If the new level playing field tax regime is properly applied, cigarettes entering Uzbekistan from outside the CIS should in any event reach the market at a price disadvantage to locally manufactured cigarettes owing to higher ex-factory prices, transportation costs, etc.”491

Moreover, the imposition of protective import duties had not been totally abandoned, but just delayed until the market was predominantly supplied by domestic production as it was felt that pressing for too many anti-competitive measures was impractical,

280

and that the absence of import duties could, in the short-term, benefit BAT, itself largely reliant on imports until production increased.491,666

Share Purchase Agreement On 14th May 1994 the Share Purchase Agreement was signed,656,658 with Annex A reportedly outlining the tax reforms required.608 The agreement had to be formalised by decree, so BAT pressure continued, mediated largely through Abdoukadirov, in order to shape the decree’s excise proposals.660,704 BAT pressed Abdoukadirov on the need for equal excise rates on imported and domestically produced products, arguing that under the new 90% excise rates and a system where cigarettes taxed in another CIS country could enter free of further duty:

“[T]here would not be a market for any cigarettes manufactured in Uzbekistan. Even if a countervailing excise duty were to be applied in Uzbekistan, with such large price differentials the incentive for smuggling into Uzbekistan is enormous. Conversely an export market from Uzbekistan would be most unlikely.”706

BAT was sufficiently confident of its ability to push these changes through that even before the decree was finalised Neil Bruce-Miller, project leader for Central Asia, reported that “BAT Central Asia has gained government agreement that tax on imports will be the same as on local production.”458 Just one week later the decree was passed and, as Bruce-Miller had requested,751 specified that within one month the authorities would submit to the Cabinet of Ministers proposals to establish an equal taxation regime.658,660 The negotiating team was delighted to have secured its desired excise reforms.491

Finalising the equal taxation regime BAT moved swiftly to produce the proposals on equal taxation that the decree required660 using its contacts in the Ministry of Finance to ensure that the company’s proposals shaped the Ministry’s recommendations to the Cabinet. 707 , 708 , 709 The documents indicate that, by the end of August, equalisation of excise and VAT on imports and domestic production had been agreed and would be in force by 1st September. 710 By February 1995, excise on imports was payable, a fact since confirmed by the World Tobacco File, 711 although many imports were escaping taxation.696

281

Three additional items from BAT’s earlier taxation plans320,321,513,702 were now emphasised to Abdoukadirov and his colleague Mrs Irina Golysheva, 707,708,712 head of the Fiscal Policy Unit.666 These were a lowering of excise rates, a revision of the excise structure to benefit BAT’s brands and the need to police the excise system (to ensure excise was paid in the country of origin on imports excluded from import taxes and that import duties were paid on the rest).707,709,712 The symbiotic relationship between BAT and the Ministry of Finance was again apparent with the Ministry reliant on BAT for technical details and BAT on the Ministry for political influence.707, 712

Lowering excise rates Despite internally recognising Uzbek excise rates as very low,700 BAT had from the outset aimed to lower them.320 The March 1st decree raising the rates reignited this desire with Chris Dufty stating that “[i]mmediate steps must be taken to ensure that the new rates contained in the Ministry of Finance Decree are not gazetted [sic].”699 BAT sought to persuade Abdoukadirov that excise rates should be reduced.660 BAT suggested two rather than the previous four tax categories, reducing the 90% rate to 40% for filter cigarettes and the 34-56% rates to 25% for plain cigarettes (Table 10-1).660,700,701 Minutes of a meeting with the Ministry of Finance note that:

“Although AMA [Abdoukadirov] was receptive to proposals, he was unmoved by our view that the 90% rate should be reduced now (he is far more concerned over control procedures on application of taxes to imports). His justification of the 90% rate is that the market has established a price… and that because of the current low TTF [Tashkent Tobacco Factory] cost base, Govt. must impose a high rate to receive its share of the actual selling price and that this rate would not effect [sic] actual selling prices, which would cause social unrest. This approach, when not applied to all products, including imports, is patently unfair to BAT.”660 Abdoukadirov expected BAT to make full proposals on the excise system, and BAT used this to push their agenda,660 planning for Dufty to prepare a paper that:

“will demonstrate that a high rate (90%) will lead to large scale smuggling, whereas a reasonable level of tax applied equally to all products will meet both Government revenue and TTF marketing requirements.”660

282

By July the Ministry of Finance appeared to be on board.707,712 With Cabinet approval, the Ministry could issue orders to repeal the 90% legislation707 and BAT therefore sought to input directly to the Ministry’s correspondence with the Cabinet. 709,712 It prepared a tax proposal in both English and Russian,707,708 which seemingly had the desired effect: Abdoukadirov’s proposals to the Cabinet reflected BAT’s preferred 40%/25% rates.709 Bishop reported that:

“The visit seemingly produced very positive results, in that BAT recommendations were accepted in full…. John Selby will follow up the actual response from the Ministry of Finance next Tuesday (19/7/94). If this is in line with my suggested response then I think we can say that the Ministry of Finance is definitely co-operating with BAT. If approved by the Council of Ministers (so effectively removing the 900% impediment) then for the time being the excise issue should not hinder the deal completion.”707 By September, all excise reforms were apparently resolved to BAT’s satisfaction713 and by February 1995 excise rates were lowered to the levels requested by BAT.696

Achieving detailed tax reforms on ad valorem and specific tax rates Since the government was relying on cigarette excise duties to provide a substantial portion of its revenues, BAT promoted various excise scenarios on the basis of their claimed revenue-earning potential. Documents indicate, however, that such scenarios were actually developed with BAT’s marketing and profit needs uppermost.714,715 Selby alluded to the degree of deception involved in such lobbying efforts:

“This may be difficult to argue as it does not give more revenue and they will smell a rat. I must keep their present confidence that I am trying to be objective in helping them.” 715 Based on marketing and production plans,700,702 BAT sought a structure that would benefit its brands at the expense of its competitors. Thus the inclusion of an ad valorem element would disadvantage Philip Morris’ principal brand, Marlboro, while a specific element would combat the cheap products being dumped on the market. Having successfully achieved equal taxation on imports and domestic products, BAT pressed for reductions or exemptions for its own products from both forms of excise that would further advantage its domestic production:

283

“1. The principle is to hit Marlboro by keeping the ad valorem system at around the present level for imports. Naturally I will try for a reduction on domestic rates but I can see the principle of differential tax being applied when they introduce Import Duties as from 1.7.95 … 2. As so much of the import market is at the bottom end and is already adversely impacting upon our pricing levels we are advocating a ‘de-minus’ [sic] specific rate to ensure the cheap produce being dumped here is hit. Taking marketing’s view on what they believe the market can absorb at the bottom end in terms of pricing we have come up with $3.50 per mille for filters and $1.20 per mille for plains. For ASTRA, our principle brand for the time being, this should put up the retail price, in real terms from Soms 2.50 to Soms 4.00 per pack [$0.08 to $0.13 based on conversion rates at that time] without any additional tax on us on the de-minimus. Ideally the minimum would not apply to us but assuming it will, then the Som 4 price just reaches the minimum limit thereby ensuring any under invoicing/dumping will be caught at the price level Marketing is happy with.”716 [emphasis added] In February BAT hurriedly presented its proposals to Abdoukadirov for immediate submission to the Cabinet.714 They stressed that low government revenues were attributable to tax evasion on imports (as the present system could not be adequately policed) and the high proportion of cheap cigarettes (largely imports) in the market.696 BAT’s interests were also served by its failure to inform the government that raising taxes would raise revenue,327 rather arguing that increasing taxation too quickly could cause the market to shrink:

“To attempt extracting significantly higher revenues too fast will only cause the cigarette market to shrink due to the limited purchasing power of the consumer. Excessive taxation only encourages evasion which can never be eliminated completely. A shrinking market would only negate Government actions to increase revenue. The first priority must, therefore, be to concentrate on a system that will enable the highest success in collecting taxes and limits the extent of untaxed products in the market. With just one domestic manufacturer in Uzbekistan the emphasis of the system must be directed at controlling imports.”696

284

BAT’s recommendations focused on four main areas.696 First, retaining the BAT devised ad valorem system of 40% on filtered and 25% on plain cigarettes on the grounds that “higher taxes [would be] received on higher priced cigarettes, particularly the high priced international brands”, presumably (although not stated), Marlboro.696 Second, adding a minimum tax level on imported cigarettes to both ensure that “acceptable revenues are always received” and “provide some decree of protection to domestic manufacturers from underpricing and dumping.”696 Third, the introduction of an import duty of between 15% and 25% to provide “further encouragement to all local industries”.696 The fourth recommendation was the introduction of a tax stamp or marker system, from which BAT again sought exemption.

The expected impact on prices appears to directly benefit BAT’s planned brand strategy, notably its initial primary focus on the filterless brand Astra.696,716

An ‘effective’ policing system Meanwhile BAT had begun to press firmly for the collection and policing of taxation on imports660,707,708 hoping to use the paper it was producing for Abdoukadirov to push this agenda:660

“The note will also stress the vital importance to both Government revenue and the commercial viability of TTF of the vigorous policing of tax collection on imports. The concept of practical application of control procedures is giving M. of F. most difficulties at this time and our assistance is actively sought. It is worth noting that M. of F., following pressure from IMF, have already decided to impose excise on imports of alcohol and tobacco products from 1 July 1994, so control procedures must be introduced with or without our joint venture.” 660 BAT’s response to this request for assistance again incorporated proposals to its own commercial advantage.696 It wanted to be excluded from the controls for the first twelve to eighteen months, during which time BAT would be largely reliant on imports. Chapter 6 identified BAT’s use of smuggling as a market entry strategy, a tactic subsequently observed elsewhere. 717 Combined with documents indicating BAT’s extensive “transit business” (an established company euphemism for smuggling718) in Central Asia,719 it is probable that the exclusion aimed to facilitate BAT’s involvement in smuggling.

285

However, once it had commenced manufacturing international brands locally, BAT would want inclusion to prevent local consumers recognising the difference between domestically produced and imported brands, since American or European imports were perceived more favourably:440

“As excise tax on locally produced cigarettes would be unchanged it is arguable whether tax stamps are needed at all for the domestic manufacturer. In the longer term when production of international brands commences locally it would not be desirable for such brands to be differentiated from the same brands produced overseas. However, in the short term the domestic manufacturer has only just started to refurbish it’s [sic] Tashkent factory. The application of tax stamps to packets requires the necessary equipment which is being planned,…… A period of 12 to 18 months would therefore be required before stamps could be applied to locally produced cigarettes.”696

One month later, it appears that BAT was given exactly what it had been aiming for on both the policing and new excise systems:

“Irena Golyshava has called to say that the President has given his verbal agreement to a banderole system [a hand written note says “I think they mean a tax marker system!”]. I believe this to be only the first step as Bahkrom Ibragimov (Customs Head) has been told to start up the various committees for implementation. Naturally I am making myself available to give further advice. Incidentally I believe the trip paid other dividends. We have been able to completely reorganise the calculation and payment of taxes exactly as we wanted.”497 BAT’s subsequent corporate social responsibility report on Uzbekistan indicates that tax stamps were introduced in 1996,685 and external sources confirm that duty stamps are now in place.685 Moreover, as BAT had desired,696 licensing is now required for import and export, wholesale and distribution and retail.120

Post script According to WHO data the price of cigarettes in Uzbekistan is now the lowest of all countries in the WHO Europe region, including those with which it is economically comparable. The retail price of the most popular or cheapest local brand of cigarettes in Uzbekistan is $0.01 for 20 cigarettes, with the next lowest price, that in Russia, ten

286

times higher at $0.1, or three times higher when compared on purchasing power parity.120 Given the marked disparity between the excise rates BAT successfully imposed and World Bank recommended rates of two-thirds to four-fifths of the retail price of cigarettes,327 the low prices are not surprising. Although, as indicated in Chapter 1, there are no accurate recent data on smoking prevalence, media reports and tobacco industry monitors suggest consumption720 and sales711, respectively, are increasing. It seems likely therefore that the low prices, along with the marked increases in advertising outlined in the following chapter, must have underpinned these increases.

DISCUSSION BAT is demonstrated to have thoroughly redesigned the tobacco taxation system in Uzbekistan to advance its commercial objectives. Through careful use of high level contacts, the exploitation of their naivety and the provision of half truths BAT was able to secure the introduction of excise and VAT on imports, a significant, 50%, reduction in excise on cigarettes, the design of an excise system to benefit its brands and disadvantage those of its competitors and the introduction of a tax stamp system. The only one of its original aims not achieved by 1995, when these documents end, was the introduction of import duties although BAT was still hopeful that this would occur.

These findings therefore indicate that TTC investment jeopardises the chances of implementing an effective excise regime – essential if the predicted price falls associated with privatisation, with their negative impacts on public health, are to be countered. They also suggest that if effective tobacco control policies, including increases in excise, are to implemented, this needs to be done before privatisation, if it is to have any chance of success. This is substantiated by evidence in BAT’s own corporate social responsibility report that it continues to work closely with the government on excise issues, providing the Ministry of Finance “with all necessary information for effective decision-making”, 685 a fact confirmed by my contacts.

These findings are supported by evidence that BAT’s influence on taxation policy in Uzbekistan is unlikely to be an isolated occurrence. BAT’s general strategy documents outlined above indicate that BAT aims to influence taxation policy wherever possible694,695 and to secure favourable excise reform wherever in the FSU it sought to invest. In Ukraine, where BAT successfully acquired the Prilucky

287

tobacco factory,408,421,721 it lobbied for a lower tax structure,722 despite its admission that cigarettes were “extremely cheap”,408 for the introduction of excise on plain cigarettes (previously untaxed) and for a specific element to accompany the ad valorem system.408, 723,722 Considerable reductions in excise rates during the period 1993 to 1995 suggest BAT was successful and predictably resulted in a significant fall in state revenues.556 In Belarus, where BAT attempted to establish a joint venture, documents indicate that a key investment condition was the implementation of favourable tax systems.556 When interested in acquiring the Bishkek tobacco factory in Kyrgyzstan, BAT was intensively involved in efforts to influence the excise structure, with its investment conditions including a reduction in excise on domestically produced cigarettes, combined with an increase in excise on imported cigarettes and a number of other tax and excise exemptions that would protect its domestic position. 518,724,725,726,727,728,729 Further evidence, including claims that the governments of the Baltic states contravened promises of favourable taxation rates, and documentary evidence from Hungary, suggests that other tobacco companies also expect excise concessions when investing. (see Chapter 2),219,280,730,731 Moreover four other tobacco transnationals are sponsors of the International Tax and Investment Center (ITIC, http://www.iticnet.org/). This claims to be an independent non-profit research and education foundation which has “developed trusted advisory relationships with key, senior-level policymakers” in the CIS which provide ITIC and its sponsors “a seat at the policymaking table.” Three of the reports listed on the ITIC website concern cigarette taxation and all three present a uniquely one-sided viewpoint whilst simultaneously claiming to be independent. These findings are particularly important in the context of BAT’s efforts to encourage FSU governments to accept its investment by alleging that BAT involvement would increase excise revenues (see Chapter 7) and the IMF’s efforts to promote privatisation on the basis of its economic benefits. Here we see that the opposite may occur - as a result of BAT’s dedicated lobbying, excise rates were reduced. Government revenues will have been further depleted through BAT’s involvement in smuggling (Chapter 6) and the numerous other fiscal privileges and exemptions it negotiated (Chapter 9). The latter were detailed in Chapter 9, which established the extent of BAT’s anticompetitive actions. The tax concessions detailed here served to further minimise

288

competition in the market. This approach and BAT’s extensive efforts to alter the taxation laws are clearly at odds with the OECD’s business guidelines that discourage “seeking or accepting exemptions not contemplated in the statutory or regulatory framework related to environmental, health, safety, labour, taxation, financial incentives, or other issues.”281 Finally, the way in which BAT redefines its approach according to the circumstances in which it finds itself and the stage in its development of a market presence also deserves comment. Chapter 9 highlighted how an industry generally known for its efforts to lower trade barriers to open up new markets was deliberately attempting to restrict trade and competition in Uzbekistan. Similarly, despite extensive evidence of BAT’s complicity in contraband, here we see the poacher turn gamekeeper with BAT seeking to establish a tax marker system to prevent its competitors smuggling into a market it considers its own. BAT aimed to exempt itself initially, presumably, as outlined above, to give itself the opportunity to benefit from smuggling. There is considerable irony in BAT lobbying for such provisions and declaring that a tax stamp system “has proved very effective in aiding the collection of taxes”,696 when the TTCs have aggressively undermined the development of effective international measures via WHO’s Framework Convention on Tobacco Control 732 and campaigned specifically against such measures, as in Hungary.772

In summary, these findings suggest that privatisation and TTC entry are antithetical to the implementation of effective tobacco excise policies. As such they constitute a major threat to tobacco control and thus to public health. They highlight that if privatisation goes ahead, effective excise policies need to be implemented beforehand and to be based on the advice of external independent tobacco control experts, not the TTCs or anyone acting on their behalf.

289

CHAPTER 11 “Unless Health Decree 30 is amended satisfactorily it will not be possible for this transaction to proceed”: British American Tobacco’s erosion of health legislation in Uzbekistan

BACKGROUND Preceding chapters have established the unprecedented changes that occurred in the tobacco industry in the former Soviet Union - the privatisation of formerly state- owned monopolies, the emergence of active cigarette marketing - and the potential impact of these changes - the increase in cigarette consumption and likely increases in smoking prevalence particularly amongst young women. This suggests that like trade liberalisation, investment liberalisation increases tobacco consumption, with the increases mediated, at least in part, through increases in marketing.98,108 Changes in Uzbekistan following the 1994 establishment of British American Tobacco’s (BAT) monopoly appear to mirror those seen elsewhere - although there are no accurate data on prevalence, consumption has, according to Uzbek media reports, increased by 7% to 8% annually, primarily amongst young people,720 and cigarette sales have risen by 50.5% between 1990 and 1996.711

Just as taxation provides the means to control the fall in prices that investment or trade liberalisation tend to herald (Chapter 10), so the implementation of tobacco control policies, particularly a comprehensive ban on marketing covering both direct and indirect advertising, would help control the increase in marketing. It was therefore of major concern that, as outlined in Chapter 9, my preliminary analysis of the documents indicated that whilst completing its deal, BAT learned that the Chief Sanitary Doctor of the Republic of Uzbekistan, Mr TI Iskandarov had issued “Health Decree 30”, a potentially highly effective piece of tobacco control legislation that, inter alia, banned tobacco advertising, smoking in public places and introduced health warnings. BAT responded aggressively, delaying completion of its investment until the decree had been replaced with a voluntary advertising code.

Until now, the only information available on the issue was a self-serving report by BAT that failed to mention the existence of the original decree, implying instead that BAT had instigated the development of the code de novo, and attempting to present the code as an example of “the company’s responsible attitude to its advertising practices.”669(Box 11-1). This chapter reveals the true story of the code’s

290

development. It outlines how marketing was seen as essential, despite BAT’s monopoly status and the industry’s claim that advertising only seeks to influence brand choice, and highlights how BAT actively undermined legislation that would have accorded with international developments in tobacco control and served to protect the health of the Uzbek population.

Box 11-1 A new voluntary code … “A new voluntary code for advertising, developed jointly by BAT and the government of Uzbekistan, has been introduced in the Central Asian Republic. The document crowns almost a year of intensive work by BAT and Uzbek government agencies. It has the legal force of a government decree, which all tobacco companies operating in Uzbekistan will be obliged to comply with. Entitled The Provision of Tobacco Advertising Conduct, it sets out guidelines for tobacco advertising, in accordance with the norms adopted in a number of countries throughout the world. BAT’s key role in formulating the code, is seen by many as a reflection of the company’s responsible attitude to its advertising practices.”669

METHODS The full methodology is outlined in Chapter 5 with Chapter 9 giving specific details of the searches used to identify documents on Uzbekistan.

Once fully indexed, the documents used in this chapter were extracted from the project database by searching for documents relating to Uzbekistan, “marketing and advertising”, “regulation and legislation” and “policy influence”. A secondary search of all indexed documents on Uzbekistan was performed to ensure no relevant documents had been missed. Various materials were then used to triangulate and contextualise the data, most notably other documents, internet searches, newspaper reports, industry journals and interviews with key players as outlined in Chapter 9. Whilst these interviewees cannot be named, they served to corroborate the evidence presented in this chapter.

RESULTS The marketing environment in the early 1990s Tobacco advertising was slower to develop in Uzbekistan than its westerly neighbours. In 1993 BAT marketing expert Murray Marr described the advertising environment as

“Very primitive, ….. An occasional sight where it exists.

291

Outdoor evident in the form of a single-site, sophisticated electronic billboard alternating Kent and lucky Strike, seen on the outskirts of Tashkent ...” 403 His colleague Dean Sims found it “astonishing given the three years since independence, that there is so little brand support” for products, emphasising the remarkable opportunity this presented:

“The Uzbekistan market must be unique in the world in terms of its singularly unexploited advertising and promotional environment. Virtually no active communications whatsoever are apparent…. …..In a community starved of this sort of material for so long, trade and consumer loyalty could be quickly achieved and the difficulty/cost of establishing brands thereafter exacerbated. ….Advertising costs will be cheap enough to allow multinationals almost unrestricted market spend ….”497 Despite uncertainties over the legislative position,321,661,733BAT’s subsidiary Brown and Williamson had already started advertising, claiming to be the first international tobacco company to use television commercials and point of sale placements “to provoke consumer awareness of Kent, L. Strike, Pall Mall SL and Viceroy and stimulate demand/purchase for these trademarks”.734

Marketing plans BAT’s plans projected that between 1993 and 1999 annual cigarette consumption would increase dramatically from 1,100 to 1,300 per capita or 22 to 32 billion units (a 45% increase),616,661 while BAT would obtain an 80% market share, with annual sales exceeding 25 billion.616,661 BAT would achieve this by increasing supply735,736,737 and stimulating demand.661 Documents refer to:

“the potential level of consumption that BAT could immediately stimulate through improved product, increased domestic production, improved distribution (particularly to rural areas) and marketing activities”.737

Distribution was clearly vital with an exclusive arrangement with the state distributor seen a pre-requisite to investment and the increased availability of inexpensive cigarettes seen as key driver of the predicted increase in market size alongside population growth, increased per capita consumption as the economy improves and a “growth in incidence amongst women as cultural stigma on smoking recedes”.497,661

292

This anticipated increase in female smoking is reflected in plans to target marketing at women with documents noting how:

“Historically, local products have been too strong to attract large numbers of female smokers. Female smoking is now more socially acceptable and females can be drawn into the market via menthol offers or lighter brands…”497

In addition to women, young people, “opinion leaders” and those living in urban areas were identified as targets.661, 738

Freedom to advertise BAT’s marketing objectives were predicated on the existence of an unrestricted advertising environment.616,739,740 Its financial advisers, Schroders, argued that:

“BAT would require freedom to advertise its domestic and international tobacco products both generally and at the point of sale. BAT would require an undertaking from the government not to impose restrictions on the advertising of tobacco products for a period of (seven) years from the agreement to invest” 321 Although freedom to advertise was described as “desirable but not mandatory”, its importance as a pre-requisite to investment rose once its continuation was threatened.321,497

Poor understanding of health issues Documents suggest that the health consequences of tobacco were very poorly understood in Uzbekistan, underlining the need for effective tobacco control measures.406,497 Consumer research showed that some believed that certain filters were “reliable/safe to one’s health”.406 One marketing report notes:

“There is no real understanding of tar/nicotine levels and subsequently no understanding of the positioning of “lights” variant offers….. There is little awareness of the smoking and health issues.”497

Health Decree 30 Given the comparatively swift progress of its investment negotiations (see Chapter 9) BAT was shocked by its discovery, on 25th August 1994, that the Ministry of Health had issued a tobacco control decree in July, entitled Health Decree 30.664,741,742 Documents suggest that BAT had already succeeded in reversing664 a decree banning

293

street advertising in the capital Tashkent,743 but Schroders’ William Wells, faxing the news to BAT, suspected that the Ministry of Health would prove “an altogether more difficult animal with which to deal”.664

Wells described Health Decree 30 as covering a number of “highly sensitive issues”,664 presumably encompassing its recognition of the dangers of smoking, the increase in tobacco related diseases in Uzbekistan and their high prevalence in male smokers (Table 11-1).742 The decree noted that Uzbekistan had no tobacco control laws, that, in contrast to BAT’s reports just one year earlier, tobacco advertising “goes on a large scale” undermining the effectiveness of health promotion efforts.742

The decree constituted a highly effective piece of tobacco control legislation, without precedent in the region. It banned tobacco advertising, smoking in public places, sales from unlicensed outlets, the manufacture and sale of filterless cigarettes, the sale of high tar and nicotine cigarettes (setting levels at 1mg nicotine and 15mg tar) and introduced health warnings (Table 11-1).742 Its development by Iskandarov was a remarkable achievement.

BAT’s approach BAT believed the decree represented a material adverse change under the terms of their agreement with the Uzbekistan government664,744 and immediately pursued its reversal or deferral.664, 745, 746, 747 Both Dr Sharon Boyce (then Manager Smoking Issues and previously Senior Scientific Advisor at BATCO) and Ulrich Herter (Managing Director Tobacco, BAT Industries) were primarily concerned with clauses relating to the advertising ban and the tar/nicotine reductions.608,748 For Boyse:

“the main problem areas are advertising and the proposed tar/nic ceilings… The provisions for smoking in public places, provided they permit some degree of flexibility and allow smokers to smoke, albeit in special areas, seem quite reasonable compared to what we normally have to face!”748 Within BAT the decree was variously assessed as having a “serious effect on the joint venture proposal”665 or as a “deal stopper”.749

294

Table 11-1 Health Decree 30 – the original text and BAT’s response (quoted directly) TOPIC ORIGINAL DECREE742 BAT RESPONSE IN MEETING WITH HEALTH MINISTER 746 Health impacts 1. Smoking belongs to the most widely spread bad habits and it is one of the main Mr Iskandarov assumed that BAT accepted this as indisputable fact. Our reply was that the risk factors related in the development of chronic non-infectious diseases such BAT people sitting around the table were not qualified to comment. Nevertheless, our as chronic bronchitis and emphysema, lung, throat, larynx, mouth, gullet, understanding was that these were complex scientific issues and that there was still a pancreas, urinary bladder cancer, cardiovascular diseases. Smoking can controversy about smoking and health. We offered to bring our experts to explain our position essentially worsen the clinic [sic] course of such diseases as stomach and in more detail but this was not followed up on by Mr Iskandarov. duodenum [sic] ulcer, chronic gastritis, etc, strongly influence the development of the heart-crowning artery [sic] sclerosis. Smoking has a strong negative influence on the women of fertile age and on the health of their children. 2. In Uzbekistan the number of people suffering from the diseases of breathing system, including those related to smoking, increased from 1.8m people in 1991 to over 2.0m in 1993, the number of people suffering from tumours increased from 106.2 thousand up to 136.8 thousand ……blood circulation system – from 731.1 thousand up to 824.2 thousand people…. 3. Retrospective observation of the 40-57 year-old men in different cities of Uzbekistan showed that deaths caused by oncologic [sic] diseases occur 3 times more frequently, and deaths caused by the heart pathology 2.5 times more frequently among the intensive smokers compared to people who never smoke. 75% of the people suffering from the chronical [sic] bronchitis smoke or used to smoke. 4. In Uzbekistan 49% of men and 9% of women smoke. Among the 20-29-year- old-people 60% of men and 11% of women are smokers, among the 30-39- year-old-people – 53% and 15% correspondingly in the recent years the number of smoking children and teenagers tends to increase. … 5. Since recently the sales of tobacco products of foreign production frequently without documents confirming the quality of such products, have become widely spread in Uzbekistan. The Republic of Uzbekistan has no laws which restrict smoking, and the advertising of tobacco products goes on a large scale. 6. Wide advertising of tobacco products through the mass-media makes the sanitary-englighting [sic] work on the harm of smoking inefficient. 7. Almost all developed countries of the world (USA, Canada, Germany, Sweden, Norway, etc) have already taken the preventing measures for the smoking epidemy[sic]. The first results of this work show that smoking among the population can be successfully overcome. Taking into consideration the above and for the purpose of decreasing the spread of smoking, decreasing the morbidity, disability and lethality caused by the basic chronical [sic] non-infectious diseases related to smoking based on the Law of the Republic of Uzbekistan “On the State Sanitary Supervision”, I issue the decree as follows:

295

Cigarette 1. From the moment of issuance of this Decree the following shall be permanently Concerning the upper delivery targets of 1mg nicotine and 15mg tar, deliveries prohibited to [sic] the Republic of Uzbekistan: 6. This in itself will preclude the manufacture of plain cigarettes and, we believe all current 1.1 Sale of tobacco products without documents confirming their quality and with filter brands produced by TTF. In effect TTF will have to close with immediate effect…… the content of nicotine exceeding 1mg and tars exceeding 15mg in one 7. At present TTF can produce only plain cigarettes due to materials shortages, and that their cigarette……. capacity for filter cigarettes is very small. 8. Before setting targets for delivery reductions it is important to understand the delivery figures of the current domestic products. … 9. Reducing cigarette deliveries is possible but in the case of the domestic tobacco industry will require the installation of new equipment, and may also require the introduction of new tobaccos with low nicotine levels [do we have any measurements of Uzbek tobacco chemistry?] This cannot be achieved with immediate effect but can only be introduced in a phased manner.

Measurement of deliveries: It was pointed out that there are internationally accepted instruments and test procedures used to measure cigarette deliveries and that BAT would be happy to advise the government laboratories on this. This was welcomed.

Documents confirming quality: ….. If Mr Iskandarov’s need for documentation cannot be changed then BAT should at least propose a system that is manageable as the Uzbeks appear to have little idea of the practicalities. Advertising 1.2 Advertisement of the local and foreign brand tobacco products, including the 10. The decree will seriously interfere with the industry’s commercial freedom to market a advertisement through the mass media (television, radio, cinema, newspapers, legal product. magazines, etc); 11. World wide experience consistently shows that advertising bans to not reduce consumption. 12. Advertising a mature product like cigarettes is not intended to increase the overall market but to expand company market share. A ban would prevent manufacturers from informing smokers about which cigarettes are available and what their attributes are. 13. The international tobacco industry takes a responsible attitude to the marketing of its products and in many countries works with the government to agree a set of advertising standards (not targeted at young people etc.). In Russia a voluntary code has just been signed. 14. A comprehensive ban on advertising would discriminate heavily against BAT as it is a relative newcomer in the market and would be to the ultimate detriment of the domestic industry.

It was clear from discussion that Mr Iskandarov is concerned about advertising encouraging young people to smoke. BAT made it clear that this was not their intention. BAT offered to make available a copy of the voluntary advertising code in Russia and Mr Iskandarov was interested to receive this; we already have the Russian version available but

296

will not hand it over until a coordinated approach to this whole issue has been established. News filtering back via KPMG from Mr Mahsudov indicates that adoption of the Russian code until an Uzbek code can be formulated may be the way forward. Smoking in 1.3 Smoking in the public places, on the transport, at the health care institutions, BAT made the point that scientists do not agree that other peoples cigarette smoke has been public places kindergartens, schools and other institutions for children, colleges and proven to be a cause of disease. However, we recognised that cigarette smoke can be irritating universities. At other institutions and enterprises smoking should be permitted to both smokers and non smokers but believe the solution is not necessarily a smoking ban but only in specially arranged places. improved ventilation. Manufacturing 2. To ban the manufacturing of tobacco products without filters and with the content (See above under cigarette deliveries) of filterless, of nicotine exceeding 1mg and tars exceeding 15mg in one cigarette, to replace high tar/nic their manufacturing by the manufacturing of harmless cigarettes with more cigarettes perfect filters. Introduction of 3. To ban the manufacturing of tobacco products without the warning inscription on BAT made the point that there is a community expectation that cigarettes should contain health warnings the pack as follows: “The Ministry of Health Care of the Republic of health warnings. As a responsible manufacturer we respond to this expectation by placing Uzbekistan warns: smoking is dangerous for your health.” health warnings on packs of cigarettes that we produce. This in no way indicates that the company accepts that smoking has been proven to be a cause of disease. On further discussion it became clear that Mr Iskandarov was only planning that this should apply to locally manufactured cigarettes because he could not control the manufacturers of imported cigarettes. BAT informed him that this was very inconsistent, and that indeed imported cigarettes can and must be treated in the same way as domestic cigarettes. [In the Uzbek market Marlboro sells at 9 som with a warning in Uzbek and for 15 som without a warning in Uzbek. They are perceived as being different products.]

Ban on 4. To ban the sale of tobacco products in the places which are not fixed by the After much discussion it would appear that Mr Iskandarov [sic] principle concern is to unlicensed hakimates for this purpose. prevent sales of cigarettes to and by persons under the age of 18.. ….. sales To achieve this he wants to: - Remove small sellers such as tray sellers because they are often run by young people. - Only use licensed outlets And for this to be specifically aimed at tobacco products. BAT’s concern was expressed at the apparent power this gave city hakim’s to control the distribution of products. In the end no firm conclusions were reached but we suspect that in practice both sides may not be too far apart. The solution may be to propose that “cigarettes may only be sold in outlets which satisfy government guidelines” and then to specify the guidelines.

297

BAT’s initial Response News of the decree precipitated an immediate reaction within BAT. Within 24 hours the company had coordinated responses from its corporate affairs and smoking issues teams, highlighting uncertainties in the decree’s wording and proposing counter arguments.664,748, 750 Just two days after receiving notice of the decree, BAT met with Mr Mahsudov,745 from the Cabinet of Ministers. 751 Three days later Mahsudov arranged a meeting with the Ministry of Health, notably with Mr TI Iskandarov, the Chief Sanitary Doctor and author of the decree.746,752

BAT’s response addressed each section of the decree, based on key strategies outlined below (see Table 11-1 for full details), whilst repeatedly emphasising that BAT was a responsible manufacturer, working in a legal industry producing a legal product.746

1.Alleging that investment in Uzbekistan would be jeopardised 608,745,746 From its initial meeting with Mahsudov, BAT took an aggressive, even threatening approach,745 exploiting both Uzbekistan’s need to be perceived as a safe investment environment608 and President Karimov’s support:

“It was made clear to Mr Mahsudov that this [Health Decree 30] would have a substantial impact on the industry and that it would be difficult for BAT to proceed if it remained in its present form. The negative impact on foreign investors in general of individual ministries passing legislation without discussion with the industries affected, was emphasised. It was proposed that the decree be cancelled or at least withdrawn for a period of 6 months until proper discussions could take place.”745 With the Ministry of Health,752,746 BAT similarly stressed the importance of being consulted when legislation affecting its activities was proposed, as well as the support it had from the president. The Decree was portrayed as a profound threat to the local industry and to BAT’s investment:

“Unfortunately, Decree 30, as it currently stands and as we interpret it, will lead to the immediate demise of the domestic cigarette industry and will make it very difficult for BAT to continue with its planned investment in Uzbekistan which up to now has been welcomed by President Karimov and the government.” 746

298

2. Attempting to refute health impacts 741,746,748,753 BAT attempted to refute the decree’s accurate summary of the health impacts of smoking. During the meeting Iskandarov presented these impacts as undeniable, but BAT’s refutation followed Boyse’s line, suggesting an ongoing controversy in which “smoking has not been proven to actually cause” diseases748 (original emphasis in italics):

“Our reply was that the BAT people sitting around the table were not qualified to comment. Nevertheless, our understanding was that these were complex scientific issues and that there was still a controversy about smoking and health. We offered to bring our experts to explain our position in more detail ….” BAT similarly claimed that “scientists do not agree that other peoples cigarette smoke has been proven to be a cause of disease”, advocating “improved ventilation” as an alternative to a ban on smoking in public places.746

3. Denying the impact of advertising on consumption 746 It became apparent that Iskandarov intended to introduce “a very comprehensive [advertising] ban that seemed to include anything and everything mentioned to him.”746 BAT argued that:

“10. The decree will seriously interfere with the industry’s commercial freedom to market a legal product. 11. World wide experience consistently shows that advertising bans do not reduce consumption. 12. Advertising a mature product like cigarettes is not intended to increase the overall market but to expand company market share. A ban would prevent manufacturers from informing smokers about which cigarettes are available and what their attributes are.” 746 Additionally, the international tobacco industry was portrayed as both taking “a responsible attitude to the marketing of its products” and as working with governments to “agree a set of adverting standards (not targeted at young people etc.)”, as exemplified by a recent voluntary code in Russia.746 This code had in fact been developed collaboratively by TTCsand entailed only modest and ineffective restrictions (this code is also discussed briefly in Chapter 2).754

299

The initial Uzbek reaction Despite such efforts, Iskandarov rejected BAT’s suggestion that the decree be withdrawn for six months so that “an amended decree be prepared in consultation with the tobacco industry”.746 BAT therefore shifted its focus to obtaining amendments to the decree.752

BAT’s amended decree Within a few days, in preparation for the next meeting with Iskandarov, Sharon Boyce had circulated an amended decree755, 756(Table 11-2) and a highly confidential set of briefing notes justifying the changes.753,757 She noted that whilst the legal department had not cleared these particular documents, “the concepts and wording are based on previously cleared materials”.757 This and the fact that the briefing notes753 were almost identical to the document she produced748 just one day after Well’s fax arrived,664 suggest that this was not the first time BAT had responded in such a way.

BAT’s version focused on downplaying the decree’s claims about both the health impacts of smoking and the effectiveness of health interventions, and nullifying the decree’s regulatory impact.(Table 11-2)755,756 Thus the intended total advertising ban was replaced with a voluntary code to be based on the Russian voluntary code.742,746,753 Similarly, the ban on smoking in public places, which permitted smoking only in specified areas, was replaced with a ban confined to institutions dealing with health and children, specifying that elsewhere smoking areas would be provided.755,756 It is noteworthy that, despite BAT’s claims that it did not intend to encourage young people to smoke,746 the ban on smoking in colleges and universities was specifically removed, consistent with BAT’s marketing plans.497,661 Additionally, the authority issuing health warnings was to be changed from ‘The Ministry of Health Care of the Republic of Uzbekistan’ to the ‘Ministry of Health’,755,756 so facilitating exports to other markets.753 The rationale may also have been to obscure the local origin of the product, given the greater preference for international over domestically produced products.406,497,661,746

300

Table 11-2 The original742 and amended decrees 755,756 ORIGINAL DECREE742 (the text that is later changed by BAT is highlighted in bold) AMENDED DECREE (BAT changes are highlighted in bold) 755,756, 1. Smoking can essentially worsen the clinic [sic] course of such diseases as stomach and duodenum 1. Smoking has been claimed to worsen the clinic [sic] course of such diseases as stomach [sic] ulcer, chronic gastritis, etc, strongly influence the development of the heart-crowning and duodenum [sic] ulcer, chronic gastritis, etc, and to influence the development of artery [sic] sclerosis. …. the… 2. In Uzbekistan the number of people suffering from the diseases of breathing system, including 2. In Uzbekistan the number of people suffering from the diseases of breathing system, those related to smoking, increased from …. including those associated with smoking, increased from …. 4. ….There is data that tobacco smoke is more harmful to non-smokers than to smokers. 4. [Sentence removed]. 7. Almost all developed countries of the world (USA, Canada, Germany, Sweden, Norway, etc) have 7. …. … already taken the preventing measures for the smoking epidemy[sic]. The first results of this The first results of this work show that smoking among the population can be successfully work show that smoking among the population can be successfully overcome. reduced. Taking into consideration the above and for the purpose of decreasing the spread of smoking, decreasing the morbidity, disability and lethality caused by the basic chronical non-infectious diseases related to smoking based on the Law of the Republic of Uzbekistan “On the State diseases statistically associated with smoking Sanitary Supervision”,

1. From the moment of issuance of this Decree the following shall be permanently prohibited to 1. From the moment of issuance of this Decree the following shall take place in the [sic] the Republic of Uzbekistan: Republic of Uzbekistan:

1.1 Sale of tobacco products without documents confirming their quality and with the content of 1.1 The sales-weighted average tar level of cigarettes (both domestic and imported) will nicotine exceeding 1mg and tars exceeding 15mg in one cigarette……. be reduced to 20 mg over 10 years. 1.2 Advertisement of the local and foreign brand tobacco products, including the advertisement 1.2 Advertising of local and foreign brand tobacco products, including advertising through through the mass media (television, radio, cinema, newspapers, magazines, etc); the mass media (television, radio, cinema, newspapers, magazines, etc); will be restricted according to the attached code. 1.3 Smoking in the [sic] public places, on the transport, at the health care institutions, 1.3 Smoking [‘in public places’ and ‘on the transport’ removed] at health care kindergartens, schools and other institutions for children, colleges and universities. At other institutions, kindergartens, schools and other institutions for children, [colleges and institutions and enterprises smoking should be permitted only in specially arranged places. universities removed] will be prohibited. At other institutions and enterprises, and on public transport, both smoking and nonsmoking areas will be provided. 2. To ban the manufacturing of tobacco products without filters and with the content of 2. The sales weighted average tar levels of manufactured cigarettes, both domestic and nicotine exceeding 1mg and tars exceeding 15mg in one cigarette, to replace their imported, will be reduced to 20 mg over 10 years, and filter cigarettes will be manufacturing by the manufacturing of harmless cigarettes with more perfect filters. introduced into the market. 3. To ban the manufacturing of tobacco products without the warning inscription on the pack as 3. The Sale of tobacco products will be prohibited without the warning inscription on the follows: “The Ministry of Health Care of the Republic of Uzbekistan warns: smoking is pack as follows: Ministry of Health [Replaces ‘The Ministry of Health Care of the dangerous for your health.” Republic of Uzbekistan’] warns: smoking is dangerous for your health. 4. To ban the sale of tobacco products in the places which are not fixed by the hakimates for this 4. The sale of tobacco products will be banned in places which do not satisfy government purpose. guidelines aimed at protecting young people.

301

Escalating political pressure on the Ministry of Health By September, most issues regarding BAT’s proposed investment had been resolved to its satisfaction. Iskandarov, however, refused to concede on the decree608 and negotiations with the Ministry of Health were abandoned.758 BAT’s focus shifted to First Deputy Prime Minister Djurabekov, charged by President Karimov with implementing the proposed joint venture.647 BAT had seemingly established a good relationship with Djurabekov, 608 but feared that he was “having problems with the swift and effective exercise of his influence over the Ministry of Health”, leading BAT to consider an appeal to Karimov.608,759 BAT appeared confident of presidential support,608 expecting the decree to be amended satisfactorily despite intransigence by the Ministry of Health. 608, 760

An unsigned order to be issued by Djurabekov on behalf of the Cabinet of Ministers was faxed via BAT’s Tashkent office on 19th September. 761 This order, which documents suggest may have been drafted by BAT,713 required the Ministry of Health to amend Decree 30. BAT’s principal concerns were fully incorporated, with the tar/nicotine limits, the bans on smoking in public places, and provisions on filterless cigarettes and advertising all cancelled.713,761 The advertising ban was replaced with a new code761 seemingly based on, although even less restrictive than, the Russian voluntary code.754

BAT’s proposal was approved by the Cabinet of Ministers 762 who agreed that “Djurabekov will write to the Minister of Health formally requesting amendment, hopefully on the terms discussed with BAT”, with the expectation being that “the political situation will lead to a satisfactory amendment to the decree.”762

In response, the Ministry of Health reportedly offered BAT a two year exemption from the decree’s terms, 763 a compromise BAT dismissed as insufficient. This rejection triggered the direct involvement of Karimov:

“Djurabekov/Chzehen are writing to Karimov to inform him of this response and make him aware that unless the decree is suitably amended it is unlikely that BAT will invest, the Uzbek cigarette industry will collapse with the domestic market being flooded by imports, there will be a leaf farmer crisis and Uzbekistan will have its reputation as a place in which to invest very

302

materially damaged. This letter is to be delivered to Karimov’s home and it is likely that it will received [sic] prompt attention.”763 Although there were concerns about the President’s authority to amend the decree, 608 once the dispute reached presidential level its resolution appeared inevitable. BAT cited 31st October as a date when the decree would be “amended and in force” noting that further progress on the deal was conditional on this. “ 764 [original emphasis]

Within a month the deal had been completed670 with BAT transferring its first payment in November 1994671 for acquisition of a majority stake in the Uzbek tobacco industry.670

Postscript From the mid-1990s, in the absence of any effective checks, tobacco advertising in Uzbekistan became ubiquitous (see Chapter 2). 765 Unsurprisingly, tobacco consumption increased711,720 and, by 1999, BAT had achieved a market share of over 70%,211,220 not far short of its 80% target.

In 1998 a new advertising bill was introduced, amended in 2002, strengthening controls on direct advertising by apparently implementing a ban on television, radio and billboard advertising.120 Cable television, magazine, cinema and point of sale advertising are still permitted and indirect advertising remains virtually unrestricted. The bill is, however, vague and no system of enforcement is specified. Moreover, the preferred industry focus on children remains, via partial bans on distribution of samples to minors, sponsorship of events aimed at minors and advertising in magazines aimed at minors.

There are still no effective restrictions on public smoking nor requirements for health warnings on packets, ingredient disclosure or restrictions on tar or nicotine levels.120

Thus, ten years after BAT’s investment the legislation has improved little other than a partial ban on direct advertising, which can now serve only to maintain BAT’s dominant market position. Meanwhile the weak restrictions in other areas must serve to ensure that consumption is maintained and young people are attracted to the habit.

303

DISCUSSION The documents revealed here highlight firstly the massive increase in advertising that accompanied BAT’s investment. Secondly they illustrate the emphasis BAT places on both marketing and distribution in its efforts to establish demand for its products in new markets, providing further evidence of the key role such forces play in increasing cigarette consumption following privatisation. By extension therefore, tobacco control measures are essential to controlling this increase. Yet the third main finding is the negative impact BAT had on health legislation, fighting aggressively to overturn tobacco control legislation that, if effectively implemented would have protected the Uzbek population from the adverse impact of tobacco industry privatisation.

BAT’s success in overturning the bans on tobacco advertising and smoking in public places outlined here stand alongside its success in securing significant reductions in cigarette excise rates as outlined in Chapter 10. In other words, BAT removed the three most effective means of controlling tobacco consumption from the health policy arena in Uzbekistan, thereby ensuring Uzbekistan’s tobacco epidemic with its appalling health and economic impacts would reach greater proportions than would otherwise have occurred.

Boyce’s swift response and BAT’s reversal of a Tashkent street advertising ban suggests this was not the first time BAT had successfully reversed tobacco control legislation. Other documents indicate that when BAT was considering manufacturing in Kyryzstan, deal conditions included a voluntary code and agreement that no advertising restrictions would be introduced766 in addition to the wide-ranging excise reforms outlined in Chapter 10. Documents also suggest that reversal of a Soviet decree banning tobacco advertising was a precondition for the deal by RJ Reynolds and Philip Morris to import 34 billion cigarettes to the Soviet Union in the early 1990s.767

By highlighting that the considerable increase in advertising occurred prior to BAT’s investment this work underlines how tobacco control policies must not only accompany but ideally precede TTC investment. It simultaneously highlights the difficulties entailed in developing tobacco control measures in a context of industry privatisation and TTC investment, reiterating that the involvement of tobacco companies in developing legislation is antithetical to protecting health.768 BAT’s

304

behaviour in misleading the Uzbeks on a number of issues, including the health impact of smoking and its promotion of a voluntary code on advertising knowing it would be ineffective,769,770 shows clearly that investing companies cannot be trusted to work with governments to develop tobacco control measures.

In Chapter 2 I suggested that those FSU countries with major TTC investments and highly centralised one party systems of government faced the greatest challenges in implementing effective tobacco control policies. These findings support this contention: BAT’s investment accounted for over one third of total foreign direct investment into Uzbekistan between 1992 and 2000; here we see the damaging consequences for public health entailed by such influence. The Chief Sanitary Doctor, despite his best efforts, was powerless next to BAT, particularly given its close alliance with President Karimov. These documents indicate the extent to which BAT was able to exploit the centralised authoritarian character of the Karimov regime to advance its commercial objectives and undermine public health. Given the scale of human rights abuses perpetrated in Uzbekistan, such complicity is difficult to reconcile with the company’s claims of socially responsible conduct in Uzbekistan.685

In conclusion, the findings reported here support those of the earlier chapters of this thesis in highlighting the adverse public health impacts of tobacco industry privatisation. In particular they support the conclusions of the previous chapter that where tobacco industry privatisation goes ahead it should be preceded by the implementation of effective tobacco control policies and that the tobacco industry should be excluded from the policy process if such measures are to be effective. In addition, as outlined further in the following chapter, the evidence now appears sufficient to call for further changes in the approach to tobacco industry privatisation.

305

CHAPTER 12 Discussions and conclusions

METHODOLOGICAL LIMITATIONS Before examining the implications of the findings of this thesis it is necessary to consider the overall limitations of the work that has been undertaken. In this concluding chapter I will confine myself to general observations as the limitations of specific studies and methods were discussed in the individual chapters. From a personal viewpoint, the main overall limitation was not speaking Russian. Although for the vast majority of the work it was an asset to be a native English speaker, and while I work with several native Russian speakers and had access to translation facilities, in some instances it would have been very helpful to speak and read Russian.

A similar potential weakness is the broad geographical reach of the work, which meant it was impossible to immerse myself in any one country, identifying the legislation, understanding at close hand the culture and the circumstances at the time. On the other hand, this broad reach is also a strength, allowing me to examine events and patterns across a region and draw comparisons, so reaching conclusions that would not have been possible if examining one country in isolation. This is particularly important because the documents and data cover both the Soviet and post-Soviet periods and many of the documents are regional in focus, thus carving out any one country for examination would have seriously and artificially limited the analysis.

The document work is perhaps the most controversial aspect of this thesis. Although specific methodological issues were raised in advance and discussed in the individual chapters, particularly chapter 5, one key issue is worth examining more broadly here. That is the generalisability of the document findings to other countries and other TTCs. Two key questions should be considered. First, by examining selected country case studies, what conclusions can be reached about BAT’s activities elsewhere in the region? Second, by focusing only on BAT documents, to what extent can the conclusions reached be generalised to other TTCs?

The first is fairly straightforward to answer. By searching for documents on the FSU as a whole and then selecting case studies, it was possible to be sure that the case studies

306

were not extraordinary, but broadly representative of BAT’s behaviour elsewhere in the region. Indeed, as mentioned in each of the individual chapter discussions, the events that took place in Moldova and Uzbekistan mirror similar activities elsewhere.

Whether BAT’s actions can be read as “TTC” actions is more complex. I did not of course specifically search for other TTC documents (other than in Chapter 8). If I had done so, yet had still failed to establish whether a particular activity had been carried out by other TTCs, I still could not have been sure that it had or had not happened. For example, the US-based TTC documents do not detail smuggling activities. Yet BAT’s documents consistently suggest that smuggling was a practice engaged in by all TTCs (see Chapter 6). Such discrepancies are more likely to reflect the more litigious US environment and the more detailed legal review of the US-based company documents before their submission to trial lawyers as part of the discovery process. In some practices then, such as smuggling and misleading governments on their investment plans, there is sufficient evidence in the BAT documents to suggest that other TTCs also partook. There is also evidence that the TTCs collaborated in efforts to prevent advertising restrictions, for example in their collective efforts to establish the Russian voluntary code on tobacco advertising. Other academic analyses of tobacco industry documents, which have focused largely on the US-based TTCs, show that these companies engage in similar activities to those highlighted in this thesis. Although none of this other work has examined the FSU and only one former eastern-bloc country, Hungary, has been examined in detail, the findings that have emerged are broadly similar.771,772, 773 In addition, observations of TTCs activities on the ground, combined with other evidence and newspaper reports suggests that all TTCs are involved in massive marketing efforts and attempts to influence policy. For example, four other TTCs are sponsors of the International Tax and Investment Center which clearly aims to influence policy on cigarette taxation in the FSU (see Chapter 10). In totality, therefore, the evidence suggests that other TTCs engaged in similar practices to BAT. Thus, whilst it remains impossible within the confines of this project to be certain that the US-based TTCs would have engaged in exactly the same behaviour as BAT, it would seem prudent, when drawing lessons from this thesis, to proceed on the basis that they would behave similarly.

307

CONTRIBUTION TO KNOWLEDGE This thesis set out to answer a number of questions. Its findings in relation to these questions are outlined below, with emphasis given to findings that are new. Much of this work is now published (see Appendix) and has made an important contribution to the literature as I attempt to highlight.

What was the scale and nature of the TTC contribution to foreign direct investment in the FSU? This thesis represents the first effort to assess TTC contributions to FDI in the FSU and indeed to my knowledge, in any world region. Chapter 2 shows that the TTCs made major investments that within the period 1992 to the end of 2000 totalled over US$2.7 billion. These investments were focused on ten countries where the significance of their contribution to FDI varied widely from 1% (Latvia, Azerbaijan) to over 30% (Uzbekistan).

Analysis of the emergent investments (Chapter 2), alongside tobacco industry documents on investment plans (Chapter 7), suggest that investment patterns, including the number of companies investing (and thus whether a competitive or monopolised market emerged), the size and type of investment, appear to reflect a number of factors. The first is the size of the country: Russia and Ukraine, the most populous states, received the greatest investments from the largest number of tobacco companies; the far smaller Baltic States received investments from only one company. Whilst this is likely to reflect the market potential these countries represented to the TTCs, it is also of course reflective of the second factor, the size of the tobacco industry and tobacco farming infrastructure, i.e. what was on offer. Turkmenistan, for example, had no tobacco manufacturing facilities and thus received no investment. Moldova was of interest not only because of its cigarette factory but because of its leaf-growing opportunities (Chapter 8). The third is the strategic importance of the market, both in size (as detailed above) and geo-politically. For example Uzbekistan was central to BAT’s desire to dominate the central Asian market and link the Far East with Europe. Moldova was to form a base for BAT’s exports to the Caucasus (Chapters 7, 8 and 9). Finally, political factors also played a role in determining outcomes. Conflict and

308

political instability delayed but rarely totally precluded investment; so keen were the TTCs to establish themselves here (Chapter 7). Political resistance to privatisation thwarted the TTCs efforts in Moldova and Belarus. In all three Central Asian countries where the TTCs invested and where centralised one party systems are in place, monopoly positions emerged in two (Uzbekistan, Kyrgyzstan) and a de facto monopoly in the other (Kazakhstan) (Chapter 2).

The proportion of total FDI accounted for by tobacco-related investments in each country is of course determined both by the size of the tobacco investments (thus reflecting the factors outlined above) and by the size of total FDI. The latter reflects the size and number of investments in other sectors, in turn determined by the political situation, the degree of market reform and the industries available for privatisation. Chapter 2 shows that tobacco’s contribution was highest in Uzbekistan and probably Kyrgyzstan, although no figures are available for the latter, which can largely be explained by these countries having the lowest and third lowest rates, respectively, of FDI per capita of the 10 countries examined in detail. Tobacco’s contribution was lowest in the Baltics (which had the largest per capita total FDI of those countries examined, reflecting the substantial investments in other sectors) and the Caucasus (reflecting the relatively small, largely tobacco-agriculture focused investments and in Azerbaijan, substantial FDI in oil).

How have TTC investments changed the pattern of cigarette production and trade in the FSU? This thesis represents the first comprehensive effort to assess the impact that investment liberalisation has had on a tobacco market, its pattern of production, trade and consumption. The findings leave little doubt that TTC investments have totally changed patterns of cigarette production and trade. Chapter 2 shows that cigarette production capacity across all factories receiving investments tripled from 146 billion cigarettes per annum pre-investment to 416 billion post-investment. Increases vary by country, with increases as great as ten-fold seen in Kyrgyzstan. As most factories were operating well below capacity pre-investment absolute increases in production are likely to be higher than the capacity figures suggest.

309

Actual cigarette production data (Chapter 3) show that production has increased exponentially, reaching far higher levels than ever before. As expected from the capacity increases detailed above, this increase was seen almost exclusively in the countries receiving TTC investments, amongst whom production increased by 96%, compared with just 11% in those that did not receive investments.v

As a result, in the countries where they had invested, the TTC’s combined market share increased from zero to between 50-100% by the year 2000. The highest market shares reached by a single company were seen in Kazakhstan and Lithuania where Philip Morris’ shares reached 84% and 75% respectively (Chapter 2). Although TTC market shares in the countries in which they did not invest are far lower (for example, 15% in Belarus and 4.8% in Moldova), as outlined in this thesis, these are official market figures and data, for example on the high tobacco advertising expenditure in Belarus, and industry documents on smuggling, suggest that their unofficial market share in such markets is likely to be much higher.

Despite this massive increase in production, little improvement in the region’s trade figures has been seen because, as explored further below, the vast increases in production have been channelled almost entirely into local consumption. Indeed, although the initial increase in cigarette imports seen in the early 1990s has now reversed, and exports have increased through the 1990s, imports still substantially outweigh exports (Chapters 3 and 7). I would suggest, on the basis of market plans observed in the documents and outlined in industry journals, that this pattern will not now change substantially.

This thesis also shows that the rapid increase in cigarette production and the shift in output from local to international blended cigarettes introduced by the TTCs have had major implications for the region’s leaf trade. Leaf imports have increased exponentially and the regional trade deficit in tobacco leaf has increased ten-fold from 1992 to 1999 with leaf production, despite the TTCs promises, still lower than ever recorded in Soviet times. These findings are important as they pour ridicule on BAT’s efforts to present its v It should be noted that in this analysis the countries receiving investment includes only 7 of the 10 that received investment. It excludes those that received investment after 1997 (Armenia, Azerbaijan and Kyrgyzstan) as it was felt that a three year period was insufficient for the investments to have had an observable impact and these countries data therefore only appear in the total.

310

investment as offering the potential for import substitution in tobacco leaf (Chapter 7). Indeed, as Chapter 7 illustrates, BAT’s interest in tobacco leaf development was essentially self-serving – it enabled the company to sell itself to aspiring ministries of agriculture whilst providing a cheap source of local leaf for BAT’s regional manufacturing efforts. Recent reports suggest that governments were seriously misled on this issue and have been hugely disappointed by BAT’s leaf growing efforts. 558

Why and how did the TTCs respond to the opportunities posed by the opening of the FSU and how did they influence the privatisation process? The document work in this thesis, much of which has now been published (see Appendix) makes an important and original contribution to the growing literature in this field. It is the first to focus on BAT, examining its tactics in detail and the first, and to date the only, work to address the FSU. Moreover, the work on Moldova and Uzbekistan stand alongside one other paper published late last year,774 as the only analyses of TTC conduct in low income countries.

It shows that in the 1990s, expansion into new, emerging markets became a key part of BAT’s corporate strategy, stimulated by the decline in consumption, but not profitability, in western markets. BAT’s willingness to restructure and create a New Business Development Unit to co-ordinate investment there illustrates the importance BAT attributed to the FSU. The marked competition for assets and substantial investments by all TTCs indicates that other companies afforded it similar priority.

A number of factors underpinned BAT’s interest in the NIS. Some were generic to most new markets - their large size, the potential to expand sales to women, the high rates of population growth and young population structures (the last two being most relevant to the central Asian states). Others were specific to the region - the shortage of cigarettes and vast un-met demand, the very high consumption rates (due to male smoking), and the opportunity to smuggle to China, which, once the FSU had opened, remained the last major frontier to be conquered.

BAT developed a staged approach to penetrating these new markets, not dissimilar to that witnessed elsewhere,15,16,92 although this thesis provides the first document-based evidence of such an approach. The initial focus was on imports as this would avoid

311

substantial risk whilst ensuring brand presence. Local manufacturing would follow, initially through licensing and then joint ventures.

While there have been previous allegations of TTC involvement in smuggling as a key market entry tactic, my analysis confirms its use and establishes the extent and importance of its contribution. Smuggling formed a key part of BAT’s efforts to establish imports and documents suggest that this was the case for all TTCs. They outline how, in the early 1990s, the majority of BAT’s cigarettes were imported to the FSU illegally, highlighting the risks BAT took in terms of re-export to other markets. This is a key issue for consideration by governments, as not only does it reduce government revenue, but it undermines public health efforts to control tobacco use and weakens local cigarette companies, making them easier and cheaper to acquire.

BAT was also prepared to take risks in securing payment for imports, using counter- trade and, more controversially, encouraging governments to use aid money to pay for cigarette imports (Chapter 6).

As outlined in Chapter 1, Shepherd maintains that the main barrier to market entry for tobacco companies is consumer loyalty for existing producers’ products which must be overcome through marketing.15 Although, as outlined in Chapter 3, investors in the FSU were able to acquire access to some existing products and brands, the documents nevertheless highlight just how key this “demand creation” effort was. Imported brands were supported with substantial marketing in order to underpin any subsequent local manufacturing efforts. Concern about the potential advent of advertising restrictions led to massive and rapid exploitation of the marketing opportunities available and to efforts to promote the company image lest this would have to be relied on in the future. Indeed my work suggests that, to some extent, brand and company promotions were used interchangeably to promote both the company and its brands.

Marketing documents for Russia, Moldova and Uzbekistan show remarkable similarity not only in the importance clearly attributed to marketing, but in its specific targets (Chapters 6, 8 and 10). The aim was to exploit the nascent desire for all things western and to focus marketing and distribution efforts on young people, opinion leaders, women and those living in urban areas. Importantly, and in contrast to the TTC’s public

312

contentions, the documents establish that BAT expected these efforts to encourage new smokers, particularly women, to take up the habit, as well as increasing consumption. Distribution, often overlooked in tobacco control literature, was also clearly key to efforts to fuel consumption (Chapter 8).

Other striking similarities emerge in Moldova and Uzbekistan. Despite BAT’s desire to establish a manufacturing monopoly in both countries, to protect this monopoly from external competition, thereby establishing an 80% market share, and its ultimate success to this effect in Uzbekistan, BAT planned its marketing efforts as if working in a competitive environment. Indeed, despite the limited competition, it did market its goods heavily, largely, the documents suggest, because of its desire to “build brands in advance of possible advertising restrictions”. Although there was clearly some competition, and more than in the Soviet era other than in its final few years, BAT had largely constrained this competition in Uzbekistan. This would imply that privatised monopolies act quite differently to state monopolies, advertising where there is little need to do so. Related documents on Kenya where BAT enjoys a stable, private monopoly, suggest that although advertising may reduce when the market position is secure, it rapidly increases again once this position is threatened.775

The other document findings presented in this thesis also provide new insights into how BAT persuaded governments to privatise and accept its investment. Importantly, they show how easily the governments of the NIS were misled, particularly by inaccurate economic advice, including, for example, arguments that BAT investment would reduce smuggling when BAT itself was involved in the smuggling, that excise revenues would increase when it was likely that BAT would lobby heavily to prevent such an event and that leaf imports would reduce when, as described above, there was every likelihood they would increase. BAT’s tobacco leaf expertise was used to ingratiate the company to ministries of agriculture, largely inappropriately as outlined above and BAT also misled governments on issues relating to tobacco sector employment. Similarly misleading arguments were made by other TTCs.556 In addition, seemingly altruistic actions, which would today fall under the rubric of Corporate Social Responsibility (CSR), were all clearly intended to help sell BAT as a potential investor.

313

This thesis also provides unique findings on BAT’s efforts to influence the privatisation process. It suggests that wherever possible, and despite the intense competition from other TTCs, BAT aimed to avoid a competitive tender, recognising that this would inflate prices, and sought to establish a monopoly position in a closed deal (Chapters 7- 9). It successfully avoided tenders in Ukraine and Uzbekistan and attempted the same in Russia and Moldova. In Uzbekistan, such efforts were combined with a variety of other anti-competitive practices including exclusive dealing, absorption of potential competitors and the negotiation of a broad spectrum of measures designed to prevent market entry. In Uzbekistan BAT also successfully negotiated a broad number of investment privileges and exemptions from a variety of taxes and duties. This was not an isolated occurrence - exemptions from profit taxes were also negotiated by TTCs in Kyrgyzstan, Hungary and Ukraine as Chapter 9 outlines.

In totality, through smuggling, avoiding competitive tenders, ensuring these privileges and exemptions and reducing cigarette excise rates (see below), there can be little doubt that the TTCs seriously reduced the revenues that the FSU governments were due from their tobacco industries. It is however difficult to establish the extent to which these practices undermined the prices paid for factories in the region. Some regional tobacco control experts allege that, even where tenders occurred, the prices paid for assets in the region were far lower than expected.556 They estimate that initial investments in and around 1993 for the six privatised factories in Ukraine came to $66 million. (I estimate that at the time of these investments the combined production capacity of these six factories was approximately 48 billion cigarettes and that over the period until 2000 total investments in these factories reached over $153 million with their capacity increasing to some 65 billion). They compare this figure to the $390 million paid by Gallaher in 2000 to the American Liggett company for a factory in Moscow which, as outlined in Chapter 2 had increased its capacity to over 40 billion cigarettes and with the $230 million paid by Philip Morris for a factory in Turkey with a production of only 12 billion. They also cite a 1995 Deloitte and Touche report, admittedly TTC funded, which predicted that tobacco investments in Ukraine would, by 1999, reach over $520 million, over three times the amount actually reached.

314

The extent of BAT’s corporate misbehaviour and the ease with which it contravened both the Uzbek law on foreign investment and established standards of business practice, including the company’s own standards, is notable. Should BAT’s conduct be representative of other industries (as Chapter 10 indicates, punitive import duties were also established in Uzbekistan to help protect the new domestic car industry following joint ventures with manufacturers Daewoo and Mercedes, raising the possibility that other companies also sought favourable, anti-competitive measures), these findings suggest that TNCs may have contributed to the poor outcomes of privatisation in the FSU. Further consideration should certainly be given to this issue rather than simply viewing TNCs as victims of the chaotic circumstances that prevail. Analogously, these findings highlight the need for open and transparent tenders and lend weight to those who argued for infrastructural reforms to precede privatisation. Finally they indicate that voluntary codes of conduct, whether they cover CSR reporting or standards of business practice, are wholly inadequate.

It is also of concern that BAT’s advisers on the Uzbekistan deal (Schroders) were aware of and seemingly willing to assist BAT in its unethical practices both in its anti- competitive actions and efforts to overturn and influence legislation. A number of large accounting firms also feature in the documents, which suggest that some may have undertaken dual and potentially conflicting roles, for example advising governments on privatisation whilst also representing or working with a TTC interested in the privatisation. Given the limited number of large, international merchant banks and accounting firms, this raises concerns about their ability to be truly independent.

What impact has trade liberalisation and TTC entry had on tobacco consumption? Cigarette consumption has increased rapidly from a low in the early 1990s to reach levels considered high by international standards. With a 43% increase in per capita consumption between 1991 and 2001, consumption is now considerably higher than ever previously recorded (Chapter 3). Although more detailed country-specific analysis is limited by the likely impact of smuggling, the data suggest this increase has been seen exclusively in those countries that received TTC investments, a finding supported by the country-specific female prevalence data. The findings are also remarkably consistent

315

with BAT’s prediction of a 45% increase in cigarette consumption in Uzbekistan between 1993 and 1999 (Chapter 11).

As outlined above, this is to my knowledge the first direct and published analysis of the impact of investment liberalisation on cigarette consumption. Given that cigarettes are normal goods whose consumption rises with income, an increase of this size at a time of major economic recession is remarkable.

Unfortunately, for the reasons outlined in Chapter 3, it is beyond the scope of this thesis to use econometric techniques to assess the exact mechanisms by which this increase in consumption may have been mediated (such an assessment is also likely to be precluded by data availability). Nevertheless, the findings indicate that a number of changes contributed to the increase in consumption including the massive increase in advertising and more effective distribution systems targeted at those with previously low levels of consumption, combined with substantial reductions in excise rates and thus prices. The lack of other effective tobacco control policies and erosion of existing measures through TTC lobbying, possibly combined with the vulnerability of the population at a time of stress are also likely to have contributed. It is noteworthy that BAT’s predictions for the massive increases in consumption in Uzbekistan appear to have been based on an unrestricted marketing environment rather than on prices, suggesting that marketing may be more fundamental to the increase in consumption.

Since this work was undertaken, I have identified one other report on this issue. To date unpublished, it examines whether the entry of private cigarette producers to Turkey and Ukraine increased cigarette consumption.683 The findings are remarkably similar, identifying marked increases in consumption and similar mechanisms by which they were mediated. For Ukraine, this is not surprising as the report uses the same data to examine consumption, albeit at national level which, as I note in Chapter 3, is limited by smuggling.w For this reason, such data must be treated cautiously, but nevertheless, show a rapid increase in consumption from 1993 onwards due to the substantial increases in production, which the report identifies as occurring exclusively in the privately owned factories. In Turkey, the market environment was slightly different in w smuggling of cheap cigarettes from Russia and Moldova to Ukraine is a real issue, estimated in 1998/9 to account for 27% of Ukrainian consumption

316

that the state owned monopoly Tekel was not privatised but instead, from 1992, the government allowed private producers to produce and sell cigarettes in competition with it. Cigarette consumption increased dramatically after “privatisation” with an estimated 87% of the increase accounted for by private producers. Again, the growth in consumption was largely accounted for by the increased cigarette production which, unlike Ukraine, was seen in both the new private and existing state producers.

In Turkey, as in the FSU, this increase is remarkable as it occurred at a time of severe economic crisis, and worryingly, against a background of progressive improvements in tobacco control. However, as in the FSU, the private producers used aggressive marketing techniques that exploited loopholes in legislation and specifically targeted young people. For Turkey alone, the report was also able to examine prices. It found that real prices (adjusted for inflation) fell initially from 1994 to 1995, but were then stable from 1995 to 1999, increasing thereafter. The authors argue that whilst the initial fall could account for some of the increase in consumption, other factors must have been responsible for the increase seen after 1995. Once again this suggests a key role for marketing.

As in the FSU therefore, the private producers in Turkey appear to have captured a growing share of the existing market and created additional demand for their products. The authors suggest that other evidence showing that youth smoking has increased and the age of smoking initiation has fallen in Turkey, illustrates the additional demand from new smokers.776

What are the current patterns and determinants of smoking behaviour in the region and how might they be explained? Chapter 4 provides new and, for some countries, the first accurate data on smoking prevalence and determinants. It also provides the first data in comparative format for all eight countries. It highlights how male smoking rates are amongst the highest in the world, varying from 43% to 65% and that female smoking rates are far lower, varying from 2% to 15%. Men smoke more than women, are more likely to be nicotine dependent, and to start smoking in childhood. Age specific patterns vary by gender and country with marked increases in female smoking in younger age groups seen in only in Russia, Ukraine, Belarus and, to a lesser extent, Kazakhstan. Comparisons with previous

317

data are limited by their fragmentary nature and are essentially possible only for Russia, Ukraine and Belarus; they suggest that female smoking rates have increased whilst male smoking rates have changed little. In addition to age, social disadvantage was an important determinant of smoking in men, whilst in women, living in a large city was the strongest predictor of smoking.

Other chapters provide explanations for these findings and suggest that overall, female smoking patterns reflect the activities of the TTCs. Higher rates, particularly in young women, are observed in countries with a longer-established TTC presence. This, and the urban dominance also observed among women, would appear to reflect the TTC marketing and distribution patterns. Combined, these findings suggest that women represented a relatively untapped market segment that the TTCs effectively targeted.

How have the TTCs influenced tobacco control policies in the former Soviet Union and how can these influences be explained? Chapter 2 shows how the TTC’s entry led to the gradual erosion of the Soviet-era tobacco control measures and to a highly effective lobby against the introduction of new control measures. It suggests that the effectiveness of national tobacco control measures corresponds broadly to the nature of the political and economic transition in each country and the size of industry investment; more effective measures are seen in democratic states with smaller or no industry investments and the least effective measures in centralised one-party states with high levels of industry investment or those with limited governmental capacity.

These findings are supported by the detailed document work in Chapters 8 to 11, which show how the industry worked assiduously to undermine tobacco control. The most detailed work covers Uzbekistan and highlights how BAT, having already succeeded in reversing a local advertising ban in the capital, Tashkent, overturned Health Decree 30, a highly effective piece of legislation which, inter alia, banned cigarette advertising and smoking in public places (Chapter 11). It had the decree replaced with another, designed by BAT to be ineffective, and in which the advertising ban was specifically replaced with a voluntary code, itself based on a code already implemented in Russia. To achieve this degree of policy influence BAT used its high level political contacts, threats to

318

abandon its investment and misleading advice – refuting tobacco’s negative health impacts and the impacts of advertising on cigarette consumption.

BAT also thoroughly redesigned the tobacco taxation system in Uzbekistan to advance its commercial objectives, bolster its monopoly position and reduce competition. It secured a significant, 50% reduction in excise on cigarettes, the design of an excise system to benefit its brands and disadvantage those of its competitors and the introduction of a tax stamp system from which it hoped to be exempt, almost certainly to facilitate its established practice of cigarette smuggling until its local production output had increased (Chapter 10).

These must serve as some of the most egregious examples of corporate influence over on public policy. They could also accurately be described as examples of “regulatory capture”, where closeness between the regulator and the industry to be regulated leads to ineffective regulation.777 As witnessed in Uzbekistan, this typically occurs through a combination of the industry’s active efforts to influence policy and limited government capacity, the latter more prevalent in lower income countries777,778 and illustrated most profoundly here by the tobacco taxation example. Ultimately BAT ended up in the enviable position of being both author and subject of the regulation in Uzbekistan. Worringly, this situation, certainly with regard to taxation policy, appears to be ongoing with BAT’s links to the Ministry of Finance now seemingly deeply entrenched.685

Although BAT’s uniquely high contribution to FDI in Uzbekistan could be used to argue that the Uzbek experience may be unique, as outlined in Chapters 8, 10 and 11, this is unlikely to be the case, although it may of course be an extreme example. For example, similar, ineffective voluntary codes on advertising were implemented in Russia and Ukraine by the TTCs acting collaboratively and were planned by BAT for Moldova and Kyrgyzstan when it was considering investing in those countries (Chapters 8 and 11). BAT strategy documents, the presence of four TTCs as founding members of the International Tax and Investment Center, and documented attempts by various TTCs to influence excise policy in Ukraine, Belarus, Kyrgyzstan and the Baltic states, suggest that efforts to influence excise policy are also widespread (Chapter 10). Other documentary evidence shows that similar processes were followed in Hungary where

319

Philip Morris and BAT evaded and violated the strong advertising legislation that was initially in place and then worked collectively to replace it with far weaker legislation, again similar to the voluntary code seen in Uzbekistan.773 Pricing was seen as particularly important in Hungary given the advertising restrictions initially in place and Philip Morris and BAT worked tirelessly to influence the tax structure to advantage their brands and boost sales, with Philip Morris at one stage price discounting to gain market share.772

It is worth exploring, therefore, what underpins this ability to influence public health policy. The lack of governmental capacity, including inexperience in tobacco control and in dealing with the aggressive lobbying tactics of transnational corporations undoubtedly plays a part, underlining the need for external support, which is considered further below. The relative powerlessness of the health ministry against the industry and its allies in other ministries (see for example Chapter 11) may also be key. Other factors are the TTC’s significant contribution to FDI, their ability to take over existing state monopolies in all but the largest countries and the centralised, corruptible systems of government present in some countries. These problems would likely be common to many low income countries, serving to highlight the ease with which TTCs could potentially influence policy. But it is also clear that the entry of the TTCs at a time of major political and economic change left the FSU particularly vulnerable to their influence; political activity was focused on state-building, development of constitutions and economic reform and thus tobacco control could be afforded little priority. To compound these problems, there was little if any resistance to the industry’s efforts from outside government. This can be explained by the lack of a developed civil society and the relatively poor understanding of the health impacts of tobacco and of effective tobacco control measures amongst public health or medical elites, who would traditionally be expected to lobby for tobacco control measures. But it also appears that there was little external support for or advice on tobacco control at a time when it was clearly much needed (Chapter 7). The World Health Organization (WHO) responded as best it was able with very limited resources, only obtaining additional funding that it targeted exclusively to the region from 1995 onwards (personal correspondence Neil Collishaw, August 2005). But the IMF and World Bank were also noticeably absent

320

from the debates over tobacco excise taking place in Uzbekistan. This highlights a weakness in the international commmunity’s ability to respond to what could reasonably be deemed a public health crisis.

Do investment liberalisation and tobacco industry privatisation pose dangers to tobacco control and health? This thesis shows, just as economic theory would predict, that investment liberalisation and tobacco industry privatisation pose major threats to public health: they tend to increase cigarette consumption and raise smoking prevalence in selected population subgroups. This appears to occur particularly as a result of massive increases in marketing, including marketing targeted specifically at groups with previously low rates of smoking, in addition to improvements in distribution and potential price reductions. Furthermore, investment liberalisation and tobacco industry privatisation lead to serious challenges to the implementation of effective tobacco control policies, thereby worsening and prolonging the negative impacts on public health.

It also highlights that the supposed benefits of privatisation (other than the increased choice and availability of cigarettes) – foreign investment flows and increased business to associated industries such as media and advertising industries have been lower than expected. The TTCs reduced potential government revenues in a number of ways – by avoiding competitive tendering, negotiating a wide variety of tax holidays and investment incentives, smuggling cigarettes and reducing cigarette excise rates. The documents also suggest that much of the spending on advertising went to industries with western headquarters, because local services were so underdeveloped.

Furthermore, given growing evidence that tobacco control is good for a country’s economy,327 the increase in consumption with the higher future burden of ill health it entails, is likely to have negative economic consequences. This negative societal impact will be compounded by the generally worse conditions provided to employees in the private versus state owned industries. Reports from Ukraine, for example, suggest that working conditions in one factory were so bad that workers went on strike, a highly unusual occurrence, that the kindergarten for employees’ children was closed, the construction of employee apartments was stopped and hours were increased without extra pay.556,683 The British Helsinki Human Rights Group visited tobacco farmers in

321

Uzbekistan and reported on their appalling plight, subsisting on livestock feed. Since BAT took over, the farmers, previously considered prosperous, have become impoverished. In BHHRG’s words BAT “appears to be exploiting local Uzbek farmers in what amounts to de facto slave labour” and due to BAT’s status as monopoly purchasers of tobacco leaf farmers “are at the mercy of BAT when it comes to the price”.600

The market structures that emerged through privatisation differ between countries in the region. In Uzbekistan, for example, a government monopoly was replaced with a private monopoly wherein competition was seriously constrained. In Russia and Ukraine, by contrast, more competitive markets were established with the Ukrainian government retaining some small cigarette factories so that private and state-owned industry competed, although the government’s market share is now tiny. It is important to note that the increases in marketing were seen even where competition was limited, implying that it may be TTC behaviour, rather than the market environment that, at least to some extent, determines the degree of marketing. In other words economic theory can only provide a partial explanation for the changes in marketing and distribution, the rest is more simply explained by the replacement of state with private, TTC-ownership, and the very different behaviours of such companies.

This thesis also highlights differences in behaviours between these two groups of companies in relation to policy influence with TTCs shown to be more likely to seek to actively influence tobacco control policies and to do so successfully, as previously observed elsewhere.16 Unlike state-owned companies, privately owned companies are required to maximise profits and returns to shareholders and will therefore oppose anything that could threaten these profits. TTCs also work on a global basis and will oppose policies that could threaten their global profitability, including for example measures that may have relatively little impact in a single market but could set a global precedent for tobacco control as occurred with plans for ingredients disclosure in Thailand.779

The very different behaviours of these companies is a crucial finding of this thesis because it indicates the importance of supply side factors in tobacco control. TTCs

322

advertise where state owned monopolies did not. TTCs strive to create demand for their products using targeted marketing to encourage population sub-groups that state-owned companies had not previously tapped to take up smoking. TTCs work assiduously to overturn unfavourable legislation and create favourable legislation in ways that state owned companies did not. In other words, supply side factors, most notably, who is supplying the tobacco products, are important.

This contravenes the current consensus on tobacco control, that compared to measures aimed to reduce demand for tobacco, measures that act on the supply side are relatively unimportant. Such an approach has been enshrined in tobacco control strategies recommended by the WHO and World Bank109,689 and more importantly in the WHO’s Framework Convention on Tobacco Control (FCTC), the world’s first international health treaty, 780 which is set to provide the future trajectory for tobacco control particularly in low and middle income countries. My findings suggest that the current demand-side approach is restrictive and would seem to ignore the particular threats currently being experienced by many low and middle income countries.

If so, how might these dangers be mitigated? One obvious answer is to avoid tobacco industry privatisation. It could be argued, alongside arguments that tobacco should be excluded from trade agreements, that tobacco should also be excluded from investment agreements and the list of state owned enterprises that are recommended for privatisation. To consider this further, it is necessary to review the two most consistent and rational arguments in favour of tobacco industry privatisation.

The first is the conflict of interest argument – that government ownership of a tobacco industry provides incentives to the government to increase sales and disincentives to impose tobacco control measures and that privatisation therefore offers opportunities to improve tobacco control (personal correspondence Peter Heller, IMF, October 2005). In support of this position, they cite the Polish experience, where privatisation and acquisition of previously state owned companies by TTCs was accompanied by improvements in tobacco control.107 This indeed was the case, but does not imply that the former led directly to the latter as a result of the change in ownership. Indeed, the

323

Polish case can equally be explained by the presence of informed and committed public health experts with international contacts and experience and concern at the increase in smoking that followed privatisation. Similar changes were seen in Thailand and Taiwan where TTC entry acted as a spur for tobacco control but where there was no change in ownership. Moreover, the Polish experience appears to be fairly unique within the context of tobacco industry privatisation. This thesis shows that throughout the FSU, privatisation has been bad for tobacco control. The same has been seen through much of the rest of CEE771,772,773 and throughout South East Europe.781

Instead, I would suggest that such conflicts of interest are largely theoretical and that the barriers to tobacco control are far greater where the TTCs have an active presence, particularly where this presence involves investment. This is because, as outlined above, TTCs behave differently to state owned companies being far more likely to attempt and to successfully influence policy.

The second argument for privatisation is that tobacco industry privatisation, like privatisation of any industry, will improve efficiency and help address macro-economic problems (personal correspondence Peter Heller IMF, October 2005). The fundamental flaw in this argument is that when a product is uniquely damaging to health, as tobacco is, any efficiency gains, including increased output and lower prices, will lead to major public health impacts that in turn have negative economic consequences. Moreover, the TTC behaviour documented in this thesis illustrates that other predicted macroeconomic benefits may not be forthcoming.

Instead, I would argue that it is possible for tobacco industry restructuring to be achieved without privatisation. For example efficiency could be improved through better technology, thereby reducing production costs. If these reduced costs were not transferred to the consumer in an attempt to drive sales, government revenues could increase without the negative public health impacts. Although there may be little incentive to do this in non-competitive market, it is at least theoretically possible. Responsibility for the running of the state owned monopoly – at maximum profitability without harming public health - could be transferred to a third party at arm’s length from government to help prevent any potential conflict of interest.

324

Similar arguments, reinforcing my calls for a supply-side approach to tobacco control have very recently been put forward by leading Canadian tobacco control experts.782,783 They argue that private ownership of tobacco companies seriously constrains tobacco control because such organisations will always try to defeat, weaken and violate tobacco control measures in their drive for profit and that only by transferring ownership to non- profit enterprises with a public health mandate can further significant progress in tobacco control be made.

However, given the current economic climate and focus on privatisation, it is incumbent upon us to examine how public health can best be protected within the context of tobacco industry privatisation, whilst ensuring that privatisation reaps the appropriate economic gains for governments.

The findings presented in this thesis suggest that the implementation of effective tobacco control policies, most importantly comprehensive bans on tobacco marketing, appropriate tobacco excise policies and controls on smuggling, are essential to prevent the negative impacts of privatisation. I would add to this, bans on smoking in public places on the basis of their proven cost-effectiveness and ability to influence public perceptions of smoking. It is also evident that such polices must precede and not just accompany privatisation. This timing is important for a number of reasons. First, because advertising by TTCs predates their investment in local manufacturing as they attempt to establish demand for the products they plan to manufacture. Second, because TTCs will make the development of favourable policies and the erosion of unfavourable ones a condition of their investment. Third, because this sequencing offers the only possible opportunity to prevent the predicted increase in consumption.

Experience and evidence from elsewhere, however, suggests that the implementation of tobacco control policies may be insufficient to prevent the increases in consumption not least because, however comprehensive the measures seem, the TTCs exploit loopholes in the legislation, simply ignore it or work to have it overturned. For example, the increases in consumption seen in Thailand and Turkey following trade liberalisation occurred against a background of fairly comprehensive tobacco control measures. In Thailand a ban on all forms of cigarette advertising was introduced in 1989, prior to the

325

1990 agreement that allowed cigarette imports from 1991 onwards, and was followed in 1992 by more comprehensive legislation including improvements in health warnings and smoke-free provisions and from 1993 onwards by regular tax increases.98,103 Although the increases in consumption were undoubtedly contained by the legislation, they were nevertheless increases and the TTCs fairly actively circumvented legislation, particularly in the early years (personal correspondence Ross Mackenzie July 2005) whilst also price discounting (further details in Chapter 1 section 2.4.2).103 In Turkey, efforts to improve tobacco control began in 1991 although legislation was vetoed until 1996, two to three years after the TTCs invested, partly as a result of TTC lobbying (personal correspondence Sue Lawrence July 2005). The marked increase in consumption in Turkey despite the economic crisis has been described above.683 In Hungary, a comprehensive advertising ban predated TTC entry by many years but was systematically violated by the TTCs and then, five years after their entry, replaced with a voluntary code as a result of TTC lobbying.773

In addition, the reality of implementing such policies prior to privatisation is fraught with difficulties: markets do not just suddenly and secretly “open” but changes occur (the fall of communism, economic decline necessitating loans and IFO intervention, trade agreements etc.) that predict market opening. Thus, although ideally plans to privatise should be remain confidential until tobacco control measures have been implemented in order to prevent TTCs that are considering investing from opposing their development, achieving this is impractical.

Whilst the implementation of such policies would enable potential investors to more accurately assess the value of the enterprise they wish to purchase, the downside, in investment terms, is that they could potentially diminish TTC’s interest and the value of the enterprise. The latter however could be counterbalanced by ensuring a transparent and competitive tendering process. Moreover, given the proven detrimental impact of tobacco use on a country’s economy, the implementation of tobacco control measures should have a positive economic impact.

326

Other lessons Although largely outside the scope of this PhD, a number of broader lessons also deserve comment. In general the evidence gathered in this PhD supports the contention that the rapid speed of economic reform, in general, advantaged the TTCs whilst disadvantaging governments (Chapters 7 and 9). It of course also posed risks, as much of the literature to date has dwelt on, but above all the documents suggest that BAT exploited the opportunities posed by the absence of regulatory frameworks to such an extent that, despite the widely perceived risks of investing in the FSU, it was in fact difficult for the TTCs to fail. Importantly, such behaviour led to large amounts of foregone revenue to governments, and this thesis suggests, may have contributed to the failure of privatisation in the region, an issue that the extensive literature on this topic has largely failed to consider.

A related and fascinating finding, particularly given the TTC’s well documented efforts to open markets elsewhere using arguments for freer trade and competition, is that BAT consistently behaved in such an anti-competitive manner.

IMPLICATIONS FOR POLICY This work identifies a number of important implications for policy which are outlined below:

1. Investment liberalisation and privatisation of state-owned tobacco companies, particularly where powerful TTCs invest, is damaging to public health and the economic benefits are not clear cut. There is therefore an urgent need to review whether the privatisation of state owned tobacco industries can be justified, even for economic reasons. The IFOs should re-visit their policy on this issue, examining their rationale for privatisation and whether it is justified by the economic and health outcomes. This will require a full evaluation of the short and long-term health and economic impacts of tobacco industry privatisation taking into account the wider economic impacts of tobacco use.

327

2. Until such a study is undertaken, and until further evaluations of the impacts of tobacco industry privatisation can contribute to my findings, it would appear inappropriate to privatise state owned tobacco industries. 3. Specific rules are needed to govern trade and investment in this uniquely harmful product. Tobacco control NGOs may wish to address in more detail the issues around tobacco industry privatisation, as they have the closely related issues around tobacco trade, in order to lobby for changes to the trade rules. 4. In the meanwhile, where tobacco industry privatisation does occur, it must be done with far greater transparency given to the privatisation deal and associated agreements, so that the economic benefits to governments are maximised and the risk of deals that are not in the public interest is minimised. If necessary, an independent third party could help manage the process. However, there is a danger in the usual merchant banks and large accounting firms acting as third parties in this way as many of them have and continue to work closely with the tobacco industry. Moreover, as Chapters 9 to 11 indicate, some were prepared to collude in BAT’s questionable business practices. Extended tax holidays and other tax and investment incentives should be avoided as they may involve considerable foregone revenue for governments. 5. IFOs pressuring for privatisation and making privatisation conditional on their loans have a responsibility to at least ensure that the above occurs. 6. If tobacco industry privatisation is to occur, a number of steps should be taken to minimise the likely public health damage. These should include: a. Conducting a health impact assessment (HIA) of the proposed privatisation in order to assess the likely short and long-term health and economic impacts, to identify danger points and mitigate their impact. Although such study would be similar to that recommended in (1) above, it would be tailored to the specific market circumstances and would, for instance, differ according to the industry’s structure, the number of people involved in leaf growing and manufacture and their working conditions and benefits, as well as according to the strength of existing tobacco control legislation. Again, if the IFOs continue to recommend

328

privatisation the onus should be on them to ensure that HIAs are conducted. b. Ensuring that privatisation is preceded by effective tobacco control legislation including comprehensive advertising restrictions, effective excise policies and controls on smuggling. It is essential that advertising restrictions are comprehensive (including restrictions on the use of company logos and promotions) and enforceable, where possible preventing the TTCs from exploiting any loopholes. Once again, if the IFOs are pressing for privatisation they should take some responsibility for ensuring that such policies are implemented. c. Ensuring that tobacco control legislation includes effective and easily implemented enforcement policies, including high and fiscally significant fines for violation. d. Privatisation deals should include agreements that prevent the TTCs from rolling back legislation that has already been put in place. e. Ideally, the ability of the TTCs to exploit global or regional media opportunities to their advantage in the market offering privatisation, should also be limited. 7. If the IFOs continue to make tobacco industry privatisation a condition of their loans, it is incumbent upon them to ensure that implementation of the tobacco control measures outlined above are also made conditions of these loans so that privatisation cannot proceed without their implementation. 8. The tobacco industry should always be excluded from the policy process when designing and implementing tobacco control measures. 9. Governments planning tobacco industry privatisation should, where capacity is an issue, receive external tobacco control advice, independent of the tobacco industry. It would be useful to develop a practical system of international advice, both one-off and on-going, for the development of such measures that can be implemented at relatively short notice in a wide number of environments, thus overcoming the problems that the opening of the FSU presented for tobacco control. Key players in such a system could include the World Bank’s central

329

Health Nutrition and Population unit, the WHO, specialist NGOs and international experts who could provide technical expertise on developing and implementing effective tobacco control policies. IMF country respresentatives should have access to these groups and individuals and should consider funding their inputs early the process. 10. Governments about to open their markets to trade or investment need to be aware of the tactics used by the TTCs. Perhaps most importantly, they need to know that TTCs are likely to attempt to smuggle cigarettes both in order to establish their brands and to weaken the local firm so that it can be more easily and cheaply acquired. Every effort must therefore be made to prevent cigarette smuggling in these circumstances. 11. Information on the economics of tobacco control and the likely impacts of trade and investment liberalisation must be widely disseminated so that governments considering these options can make more fully informed decisions that are not solely driven by the TTC’s misleading advice or the IMF’s ill-considered approach to tobacco industry privatisation. Such information should be broad based covering impacts on the economy, agricultural production, trade and health. It must also be widely disseminated to all ministries and individuals likely to be involved in decision making including for example agriculture ministries that have a tendency to be overlooked by the tobacco control community whilst being actively targeted by the tobacco industry. Country-specific data on production, trade, consumption, prevalence, tobacco’s contribution to the economy and the likely impact of tax increases and other measures to reduce tobacco product consumption on government revenues, smuggling, employment, tax incidence etc could be invaluable particularly as the contribution of the tobacco industry to the economy is often exaggerated. The WHO and World Bank could play a key role in developing such data where they do not already exist. 12. Meanwhile, the countries of the FSU need to take action to improve tobacco control in order to reduce current smoking rates and prevent uptake amongst new smokers. This will require the enactment and enforcement of comprehensive

330

tobacco control policies. In those countries that have already privatised their tobacco industries without putting appropriate regulations in place, such action must be taken urgently. The FCTC will provide a useful forum through which such changes can be achieved and the WHO with its national counterparts structure will play a key role. The IFOs, particularly the World Bank, could also do more to encourage such action through policy dialogue and conditionality for financial support. 13. Prevention measures could usefully target those found to be at high risk, particularly socially vulnerable young men and women living in urbanised areas. At a minimum, efforts should be made to ensure that tobacco control policies do not increase inequalities in health by failing to reach these groups. 14. Tobaco control experts must re-think their current emphasis on demand side approaches to tobacco control and recognise that supply side issues, most notably who is supplying the cigarettes in a market, are also important. In addition, the important role that distribution systems play in helping to spread the tobacco epidemic needs to be given greater consideration. 15. Industry self-regulation clearly will not work. Self-directed CSR initiatives are ineffective and if CSR is to contribute in the manner in which it is intended, far greater thought needs to be given to how it is conducted. Similarly, it is clear that voluntary codes of conduct are inadequate and TTCs should be more closely regulated, perhaps through enforceable agreements. 16. The development of open, pluralistic governments is beneficial for tobacco control. Whilst this is clearly difficult to address, the development of non- governmental organisations is essential and the work of the Open Society Institute in the region is therefore to be encouraged.

The World Bank has been attempting to draw up good practice guidelines on tobacco industry privatisation that may help address some of the recommendations outlined

331

above (personal correspondence Joy de Beyer) but without going so far as to suggest that policies on privatisation are revisited.x

FURTHER RESEARCH This thesis identifies a number of areas for future research. Firstly, one could examine the activities of other TTCs in the region, for example, using PM and RJ Reynolds documents. However, for the reasons outlined above, the findings are unlikely to differ from those presented here. Research funding would therefore be better spent on other areas. In terms of closely related document work, further analysis of TTC marketing strategies in other markets, particularly the largest Russian market, would be useful, so too would work exploring in greater detail the links identified between the large accounting firms and the TTCs. Analysis of the TTC’s use of CSR initiatives, combined with a broader exploration of the role and the potential advantages of CSR compared with other strategies for improving industry conduct is also essential

In addition, there is a need to regularly update the smoking prevalence data and ideally provide it in a comparable format as done here. Such work could be usefully complemented by the use of documents to develop more detailed questions on smoking behaviour, including, for example, brand choices and the influences on such choices, which would enable the impact of the TTCs activities to be examined in more detail.

The collation and analysis of data on cigarette prices across the region, during a period of major economic change that included periods of hyper-inflation, the introduction of new currencies, and re-definition of national boundaries was beyond the scope of this thesis. Therefore, further work is needed to verify the impacts of trade liberalisation and tobacco industry privatisation on tobacco consumption on a country by country basis using more detailed econometric analysis to control for changes in price, incomes and advertising over this period. Such work would help elucidate more precisely the mechanism by which privatisation leads to increased consumption and would help disentangle the impact of market structures from the impact of privatisation.

x These guidelines have now been published on the web and are available at: http://siteresources.worldbank.org/HEALTHNUTRITIONANDPOPULATION/Resources/281627- 1109774792596/HNPBrief_5.pdfm (accessed 26th October 2005)

332

In addition, to date, there has been no systematic examination of the impact of tobacco industry privatisation on the macroeconomy or social welfare nor any direct evaluation of the health impacts of tobacco industry privatisation other than that produced here. Therefore, as outlined above, there is an urgent need for an accurate and broad-based evaluation of the health, economic and social impacts of tobacco industry privatisation. In the absence of such a study, a HIA approach could be used to model the health, economic and social benefits and harms of privatisation in order to inform the debate and hopefully help the IFOs develop a more evidence-based approach. If such a model were developed, it could then be adapted for use in specific circumstances to help identify and mitigate potential dangers including, for example, impacts on the workforce as well as on public health.

Linked to this, further work on potential future structures for tobacco manufacturers, expanding on the excellent work recently undertaken,782,783 could examine alternatives to the conventional dichotomy of state or privately owned companies.

Finally, I believe the tobacco industry documents represent a unique research resource that to date has only been adequately tapped by public health community. Although the documents only cover TTC conduct, such conduct must provide insights into that of other multi- and trans-national companies. On the former Soviet Union alone the documents warrant further work examining the extent to which transnational corporations may have contributed to the poor outcomes of privatisation in the FSU as Chapter 9 suggests. The potential use the business research community could make of such documents is unprecedented.

PERSONAL REFLECTION AND LEARNING I have learnt much from undertaking this PhD. As expected, I have developed methodological skills in a number of areas, most notably in document analysis where great thought had to be given to the most appropriate way of obtaining, organising and analysing the tobacco industry documents. My work has thus contributed to the body of knowledge on the methods that should be used in this emerging field of enquiry.

333

A number of broader reflections are also noteworthy. First, the complexities of conducting research on countries with different languages and cultures to one’s own cannot be underestimated and, wherever possible, the language should be learnt beforehand. The difficulties of obtaining information from countries with closed systems of government should also not be underestimated. Finally, public health research of this nature is complex, it requires one to be an expert in epidemiology, economics, business studies, document analysis and tobacco control – all highly complex fields in their own right. There is an inherent danger of becoming a jack of all trades and master of none.

Finally, on a more personal note, having twins in the middle of writing up a PhD is an ideal distraction from the task in hand!

334

REFERENCES

1 Leon D, Chenet L, Shkolnikov VM, Zakharov S, Shapiro J, Rakhmanova G, Vassin S, McKee M. Huge variation in Russian mortality rates 1984-1994: artefact, alcohol, or what? Lancet 1997; 350: 383-8.

2 Shkolnikov V, McKee M, Leon DA. Changes in life expectancy in Russia in the 1990s. Lancet 2001; 357: 917-21.

3 McKee M. The health effects of the collapse of the Soviet Union. In Poverty, inequality and health: An international perspective. Leon D, Walt G, Eds. Oxford University Press, 2001. pp 17-36.

4 McCorriston S. Recent developments on the Links between Foreign Direct Investment and Trade. July 2000. Revised version of a paper presented at the IATRC session on ‘Explaining Patters of International Trade’, New Orleans, December 1999. Available on http://www.ex.ac.uk/sobeResearch/DiscusssionPapersEcon/Econ2000/Econ0005.html, accessed 16th January 2002)

5 Taylor A, Chaloupka FJ, Guindon E, Corbett M. The impact of trade liberalisation on tobacco consumption. In Jha P, Chaloupka F (eds). Tobacco control in developing countries. Oxford: Oxford University Press, 2000.

6 Callard C, Collishaw N, Swenarchuk M. An introduction to international trade agreements and their impact on public health measures to reduce tobacco use. Physicians for a Smokefree Canada, 2001.

7 European Bureau for Action on Smoking Prevention. Test the East. The tobacco industry and Eastern Europe. Brussels: BASP, 1991

8 Action of Smoking and Health. Tobacco Explained, The truth about the tobacco industry in it's own words. London: ASH, 1998.

9 Macalister M. The $250,000,000,000 Habit. The Observer Magazine, 8th November 1992 pages 22-30.

335

10 Weissman R, White A. Needless Harm, International Monetary fund support for tobacco privatisation and for tobacco tax and tariff reduction and the cost to public health. Washington DC: Essential Action, 2002. http://www.essentialaction.org/needlessharm.pdf

11 Baris E, Waverley Brigden L, Prindiville J, Da Costa e Silva VL, Chitanondh, H, Chandiwana S. Research priorities for tobacco control in developing countries: a regional approach to a global consultative process. Tobacco Control 2000;9;217-23.

12 World Health Organisation. Confronting the epidemic: A global agenda for tobacco control research. WHO: Geneva, 1999.

13 Samet JM, Yach D, Taylor C, Becker K. Research for effective global tobacco control in the 21st century: report of a working group convened during the 10th World Conference on Tobacco or Health. Tobacco Control 1998; 7: 71-7.

14 Krasovsky K. Abusive international marketing and promotion tactics by Philip Moris and RJR Nabisco in Ukraine. In: INFACT, Global aggression, the case for world standards and bold US action challenging Phillip Morris an RJR Nabisco. New York: Apex Press, 1998.

15 Shepherd PL. Transnational corporations and the international cigarette industry. In Profits, progress and poverty. Case studies of international industries in Latin America. Newfarmer RS (ed). Notre Dame, Indiana: University of Notre Dame Press , 1985

16 Mackay J. US tobacco export to Third World: Third World War 5. J.Natl.Cancer Inst.Monogr 1992;25-8.

17 Thayer CW. Russia. London: The Sunday Times, 1960.

18 Wikipedia, the free encyclopedia. http://www.wikipedia.org/wiki/History (accessed 1 September 2003)

19 http://www.geographia.com/russia/rushis07.htm

336

20 Remington TF. Democratization and the new political order in Russia. In: Dawisha K, Parrott B eds. Democratic changes and authoritarian reactions in Russia, Ukraine, Belarus and Moldova. Cambridge: Cambridge University Press, 1997.

21 Rashid A. The resurgence of Central Asia. Islam or Nationalism? London: Zed books, 1994.

22 Kingkade WW, Arriagea EE. Mortality in the New Independent States: Patterns and Impacts. In: Bobadilla JL, Costello CA, Mitchell F (eds). Premature Death in the New Independent States. Wahington DC: National Academy Press, 1997.

23 Bobadilla JL, Costello CA, Mitchell F (eds). Premature Death in the New Independent States. Wahington DC: National Academy Press, 1997.

24 Dawisha K, Parrott B eds. Democratic changes and authoritarian reactions in Russia, Ukraine, Belarus and Moldova. Cambridge: Cambridge University Press, 1997.

25 Dawisha K, Parrott B eds. Conflict, cleavage and change in Central Asia and the Caucasus. Cambridge: Cambridge University Press, 1997

26 Brendenkamp, H. "Reforming the Soviet Economy". Finance & Development. June 1991. pp. 18-21.

27 International Monetary Fund, World Bank, European Bank of Reconstruction and Development, Organization of Economic Cooperation and Development. A Study of the Soviet Economy Vols 1 and 2. Washington DC: IMF, 1991.

28 Gould-Davies N, Woods N. Russia and the IMF. International Affairs 1999; 75: 1-22.

29 Havrylyshyn O, McGettigan D. Privatization in Transition Countries: Lessons of the First Decade. Economic Issues no. 18 Washington DC: International Monetary Fund, August 1999. http://www.imf.org/external/publc/ft/issues/issues18/index.htm (last accessed 15/1/2002).

30 Stiglitz J. Globalization and its discontents. London: Allen Lane, the Penguin Press, 2002.

337

31 Sachs, Jeffrey D., 1992, Privatization in Russia: Some lessons from eastern Europe, American Economic Review 82, 43-48.

32 Hoffman DE. The Oligarchs. Wealth and power in the new Russia. Oxford: Public Affairs Limited, Perseus Books Group, 2002.

33 Lavigne M. Ten years of transition: a review article. Communist and Post- Communist Studies 2000;33: 475-83.

34 Megginson WL, Netter J. From state to market: A survey of empirical studies on privatization. Global Equity Markets Conference, Paris, December 10- 11, 1998. http://wbln0018.worldbank.org/Network/PREM/PREMDocLib.nsf/da461c58dad5446c8 52566c9007ba34f/a71430647fc4fad98525675a006cf13f?OpenDocument (last accessed 11 April 2003)

35 Havrylyshyn O, McGettigan D. A Working Paper of the International Monetary Fund. Privatization in Transition countries: A sampling of the literature (WP/99/6). International Monetary Fund, 1999.

36 World Bank. Transition the first ten years: analysis and lessons for Eastern Europe and the Former Soviet Union. Washington DC: The World Bank, 2002.

37 Nellis J. Time to rethink privatization in transition economies. International Finance Corporation, Discussion Paper no. 38. 1999

38 Boubraki J, Cosset JC. The Financial and operating performance of newly privatized firms: evidence from developing countries. Journal of Finance 1998; 53: 1981-1110.

39 Campaign for Tobacco Free Kids. Public health and international trade Volume II: Tariffs and privatization. Washington DC: Campaign for Tobacco Free Kids, 2005.

40 Andreev EM, McKee M, Shkolnikov VM. Health expectancy in Russia: a new perspective on the health divide in Europe. Bull WHO 2003; 81: 778-788.

41 Pinto, B., Belka, M.,Krajewski, S., Transforming State Enterprises in Poland. Microeconomic Evidence on Adjustment, Brookings Papers on Economic Activity, 1993, pp. 213-270.

338

42 Kolodko GW. Transition to a market economy and sustained growth. Implications for the post-Washington consensus. Communist and Post-Communist Studies. 1999; 32: 233-61.

43 Nolte E, McKee M, Gilmore A. Morbidity and mortality in transition countries in the European context. Background paper for thematic session “Morbidity, mortality and reproductive health: facing challenges in transition countries” at the European Population Forum. Geneva, 12-14 January 2004. http://www.unece.org/ead/pau/epf/epf_backdocs.htm

44 Bobak M, Marmot M. East-West mortality divide and its potential explanations: proposed research agenda. British Medical Journal 1996; 312: 421-5.

45 Shkolnikov VM, Mesle F, Vallin F. Recent trends in life expectancy and causes of death in Russia, 1970-1993. In: Bobadilla JL, Costello CA, Mitchell F (eds). Premature Death in the New Independent States. Wahington DC: National Academy Press, 1997.

46 Shkolnikov VM, McKee M, Chervyakov VV, Kyrianov NA. Is the link between alcohol and cardiovascular death among young Russian men attributable to misclassification of acute alcohol intoxication. Evidence from the city of Izhevsk. J Epidemiol Community Health 2002; 56: 171-4.

47 Walberg, P., McKee, M., Shkolnikov, V., Chenet, L., Leon, D.A. Economic change, crime and mortality crisis in Russia: a regional analysis. British Medical Journal 1998; 317: 312-8.

48 Cockerham, W.C. The social determinants of the decline of life expectancy in Russia and Eastern Europe: a lifestyle explanation. Journal of Health and Social Behaviour 1997; 38: 117-30.

49 Chenet,L., Leon, D., McKee, M., Vassin, S.. Deaths from alcohol and violence in Moscow: socio-economic determinants. European Journal of Population 1998;14: 19-37.

339

50 Chenet, L., McKee, M., Leon,D., Shkolnikov ,V., Vassin, S. Alcohol and cardiovascular mortality in Moscow; new evidence of a causal association. Journal of Epidemiology & Community Health 1998; 52: 772-4.

51 Cornia GA & Paniccià R. The demographic impact of sudden impoverishment: Eastern Europe during the 1989-94 transition. In: UNICEF Innocenti Occasional Papers, Economic Policy Studies, No. 49. Florence: UNICEF, 1995.

52 Townsend, P., Davidson, N, Inequalities in health: The Black Report (3rd Edition); Whitehead, M. The Health divide. London: Penguin Books, 1998

53 Marmot, M.G., Shipley, J., Rose, G. Inequalites in death – specific explanations of a general pattern. Lancet 1984; 83:1003-6.

54 House, J.S., Landis, K.R., Umberson, D. Social relationships and health. Science 1998; 241: 540-4.

55 Ferrie, J.E., Shipley, M.J., Marmot, M.G., Stansfeld, S., Davey Smith, G. The health effects of major organisational change and job insecurity. Social Science & Medicine 1998;46 (2): 243-54.

56 Rosengren, A., Orth-Gomer, K., Wedel, H., Wilhelmsen, L. Stressful life events, social support, and mortality in men born in 1933. British Medical Journal 1993; 307:1102-5.

57 Bobak, M., Pikhart, H., Hertzman ,C., Rose, R., Marmot, M. (1998). Socioeconomic factors, perceived control and self-reported health in Russia. A cross- sectional survey. Social Science & Medicine 47, 269-79.

58 Shkolnikov, V.M., Leon, D.A., Adamets, S., Adreev, E., Deev, A. Educational level and adult mortality in Russia: an analysis of routine data 1979 to 1994. Social Science & Medicine 1998; 47: 357-69.

59 Kalediene, R. & Petrauskiene, J. Inequalities in life expectancy in Lithuania by level of education. Scandinavian Journal of Public Health 2000;28:4-9.

340

60 Gilmore A, McKee M, Rose R. Determinants of inequalities in self-perceived health in Ukraine: a neglected former Soviet Republic. Social Science and Medicine 2002;55:2177-88.

61 http://www.english.batrussia.ru/tobacco/history_russia/ (last accessed 21/4/03)

62 http://www.english.batrussia.ru/production/yava/history/ (last accessed 21/4/03)

63 Dragounski D. Well – this is the Russian market. World tobacco for Russia and Eastern Europe. 1998; II (August):32-46.

64 Checkov AP. On the harmful effects of tobacco. As cited in the Literature, Arts and Medicine Database. http://endeavor.med.nyu.edu/lit-med/lit-med- db/webdocs/webdescrips/chekhov1110-des-.html (last accessed 15th December 2003).

65 Hilton, M. Smoking in British Popular Culture 1800-2000. Manchester: Manchester University Press, 2000.

66 Dimitrov V. Russian factories need 150 new lines. World Tobacco for Russia and Eastern Europe 1997;II:3-18.

67 Forey B, Hamling J, Lee P, Wald N. International Smoking Statistics (second edition). Oxford: Oxford University Press, 2002.

68 Zaridze D. Dvoirin VV, Kobljakov VA, Pisklov VP. Smoking patterns in the USSR. In: Zaridze DG, Peto R (eds). Tobacco: A major international health hazard. IARC Scientific publications NO. 74. Lyon: IARC, 1986.

69 IARC. IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals To Humans. Tobacco smoking. Lyon: International Agency for Research on Cancer, 1986.

70 Cooper R. Smoking in the Soviet Union. British Medical Journal. 1982; 285: 549-51.

71 Zubritsky AN. Russian Federation: Tobacco smoking control. Tobacco: the growing epidemic. Proceedings of the Tenth World Conference on Tobacco or Health, 24-28 August 1997, Beijing, China. Lu Mackay, Niu, Peto (eds). London: Springer, 2000 pp476-7.

341

72 White S. Russia goes dry. Cambridge: Cambridge University Press, 1995.

73 Tkachenko GB, Prokhorov A. Tobacco and Health in the Russian Federation. Copenhagen: WHO Regional Office for Europe. (EUR/ICP/LVNG 030101), 1997.

74 Rupert J, Frankel G. In ex-Soviet markets, US brands took on role of capitalist liberator. Washington Post, November 19, 1996.

75 Anon. The pushers are coming! The pushers are coming! . . . To Russia. Tobacco Control Int 1992;1:9.

76 Anon. Imports- Exports of cigarettes. Tobacco Journal International 1995:2:62- 65.

77 Anon. Imports- Exports of cigarettes. Tobacco Journal International 1999; Yearbook;135-6

78 Anon. Production of Cigarettes. Tobacco Journal International 1999; Yearbook: 144-5

79 Anon. Production of Cigarettes. Tobacco Journal International; Yearbook: 128-9

80 USDA, FAS. Tobacco World Markets and Trade. Circular series FT-3-01, September 2001.

81 Piha T. Tobacco control activities of the World Health Organization in Europe. In: Richmond R (ed). Interventions for smokers. An international perspective. Maryland, Baltimore: Williams and Wilkins, 1994.

82 USDA, FAS. Tobacco: World Markets and trade. Circular Series FT-08-98, August 1998.

83 Anon. Tobacco Traveller. Russia update. Tobacco Journal International 2000;6:82-86.

84 Prokhorov AV. Getting on smokin' route 66: tobacco promotion via Russian mass media. Tobacco Control 1997;6:145-6.

342

85 Zatonski W. An overview of the challenges in tobacco problems in central and eastern Europe. In: Slama K (ed). Tobacco and Health. New York: Plenum Press, 1995, pp. 281-3.

86 Hurt, R. Smoking in Russia: what to Stalin and western tobacco companies have in common? Mayo Clin Proc 1995; 70: 1007-1011.

87 Doll R. Uncovering the effects of smoking: historical perspective. Statistical Methods in Medical Research 1998;7:87-117.

88 Cox H. The Growth of an International cigarette industry. Chapter 1; p.3-16. In: Cox H. The Global Cigarette. Origins and Evolution of British American Tobacco 1990- 1945. Oxford: Oxford University Press, 2000.

89 Cox H. Global potential. Chapter 2; p.19-45. In: Cox H. The Global Cigarette. Origins and Evolution of British American Tobacco 1990-1945. Oxford: Oxford University Press, 2000.

90 Slade J. The tobacco epidemic: lessons from history. [Review] [45 refs]. Journal of Psychoactive Drugs 1989;21:281-91.

91 The worldwide smoking epidemic. Tobacco trade, use, and control. Council on Scientific Affairs 1. JAMA 1990;263:3312-8.

92 Connolly GN. Worldwide expansion of transnational tobacco industry. Journal of the National Cancer Institute 1992;Monographs.:29-35.

93 Litterick D. Gallaher in pact to break into China market. The Telegraph. 05/09/2002 http://portal.telegraph.co.uk/money/main.jhtml;$sessionid$MTO4YPMYKNNWDQFIQ MFCFF4AVCBQYIV0?xml=%2Fmoney%2F2002%2F09%2F05%2Fcngala05.xml&_r equestid=269958 (last accessed 3/4/03)

94 Collin, J. Think Global, Smoke Local: Transnational tobacco companies and cognitive globalization. In: Globalization and Health: Case studies. Lee K (ed). London: Palgrave/Macmillan, 2001.

343

95 Crescenti M. The New Tobacco World. Tobacco Journal International 1999; March:.51-53.

96 Link E, Rossell, S. Global tobacco- where cigarette manufacturers roam. TJI 2001;6:17-20.

97 Saloojee Y and Dagli E. Tobacco industry tactics for resisting public policy on health. Bulletin of the World Health Organization 2000; 78 : 902-910

98 Chaloupka FJ, Laixuthai A. NBER Working Paper Series. US trade policy and cigarette smoking in Asia. Working Paper 5543. Cambridge, MA: National Bureau of Economic Research, April 1996.

99 Klein N. No Logo. London: Flamingo, 2000

100 Yach, D, Bettcher D. Globalisation of tobacco industry influence and new global responses. Tobacco Control 2000;9:206-16.

101 Jha P, Chaloupka F. Tobacco Control in Developing Countries. Oxford: Oxford University Press, 2000.

102 Honjo K, Kawachi I. Effects of market liberalization on smoking in Japan. Tobacco Control 2000; 9: 193-200.

103 Vateesatokit P, Hughes B, Ritthphakdee B. Thailand: winning battles, but the war’s far form over. Tobacco Control 2000; 9: 122-7.

104 Sato H, Araki S, Yokoyama K. Influence of monopoly privatization and market liberalization on smoking prevalence in Japan: trends of smoking prevalence in Japan in 1975-1995. Addiction 2000: 5: 1079-88.

105 Callard C, Chitanondh H, Weissman R. Why trade and investment liberalisation may threaten effective tobacco control efforts 1. Tob.Control 2001;10:68-70.

106 Bates C, Connolly GN, Jarvis M. Tobacco additives: cigarette engineering and nicotine addiction. http://www.ash.org.uk/html/regulation/html/additives.html

344

107 Chaloupka FJ, Nair R. Alcohol supply: domestic and international perspectives. International issues in the supply of tobacco: recent changes and implications for alcohol. Addiction 2000; 95 (supplement 4) S477-89.

108 Bettcher D, Subramaniam C, Guindon E, Perucic AM, Soll S, Grabman, G, Joossens L, WTO Secretariat, Taylor A. Confronting the tobacco epidemic in an era of trade liberalisation. Geneva: WHO, TFI, 2001.

109 Jha P, Chaloupka FJ. Curbing the epidemic, Governments and the economics of tobacco control. Washington DC: The World Bank, 1999.

110 Mackay, J. Erikson M. The Tobacco Atlas. Geneva: World Health Organization, 2002.

111 Murray CJL, Lopez AD. Annex 2: Assessing the burden of disease that can be attributed to specific risk factors. In: Ad Hoc Committee on Health Research Relating to Future Intervention Opitions. Investing in health research and development. Geneva: World Health Organization, 1996 (Document TDR/Gen/96.1)

112 Lopez AD. Epidemiologic surveillance of the tobacco epidemic. MMWR Morb.Mortal.Wkly.Rep. 1992;41 Suppl:157-66.

113 Revenko AF. Smoking in Ukraine. 3rd European Conference on Tobacco or Health. Closing the Gaps – Solidarity for Health, 20-22 June 2002, Warsaw, Poland. Pp.13

114 Health promotion challenges for countries of the former Soviet Union: results from collaboration between Estonia, Russian Karelia and Finland. Health Promotion International 1995: 10: 219-28.

115 Prattala R, Helasoja V, the Finbalt Group. Finbalt Health Monitor: feasibility of a collaborative system for monitoring health behaviour in Finland and the Baltic countries. Helsinki: National Public Health Institute 1999

116 Puska P, Helasoja V, Prattala R, Kasmel A, Klumbiene J. Health behaviour in Estonia, Finland and Lithuania 1994-1999. European Journal of Public Health 2003; 13: 11-17.

345

117 World Health Organisation Regional Office for Europe. Health for All Database, 2002. http://hfadb.who.dk/hfa/

118 Corrao MA, Guindon GE, Sharma N, Shokoohi DF, eds. Tobacco Control Country Profiles. Atlanta, GA: American Cancer Society, 2000. http://www5.who.int/tobacco/page.cfm?sid=57

119 www.tobaccofactfile.org (last accessed 14th April 2003)

120 WHO Regional Office for Europe. Tobacco Control Database. http://cisid.who.dk/tobacco/ (last accessed 25th May 2005)

121 National Tobacco Information Online System (NATIONS). http://apps.nccd.cdc.gov/nations/ (last accessed 8 December 2003)

122 Armenia. In: World Health Organisation. Tobacco or health: a global status report. Country profiles by region 1997. http://www.cdc.gov/tobacco/who/armenia.htm (last accessed 22.4.03)

123 WHO Global NCD Infobase. Preliminary country reports available at: ftp://ftp.who.int/data/NMH/NCD-Surveillance/SuRF_1/ (last accessed 27 Oct 2003)

124 Grim CE, Grim CM, Petersen JR, Li J, Tavill F, Kipshidze NN, Chawla PS. Prevalence of cardiovascular risk factors in the republic of Georgia. J Hum Hypertens,1999;13:243-247

125 Grim CE, Grim CM, Kipshidze N, Kipshidze NN, Petersen J. CVD risk factors in Eastern Europe; a rapid survey of the capital of the Republic of Georgia [abstract]. Am J Hypertens 1997;10:211A

126 Turkmenistan. In: Tobacco or Health: a Global Status Report. Geneva: WHO, 1997. http://www.cdc.gov/tobacco/who/turkmeni.htm

127 Raudsepp J, Rahu M. Smoking among schoolteachers in Estonia 1980. Scand J Soc Med 1984; 12: 49-53.

128 McKee M, Bobak M, Rose R, Shkolnikov V, Chenet L, Leon et al. Patterns of smoking in Russia. Tobacco Control 1998;7:22-6.

346

129 Gilmore A, McKee M, Telishevska M, Rose R. Smoking in Ukraine: epidemiology and determinants. Preventive Medicine 2001;33:453-61.

130 Gilmore A, McKee M, Rose R. Smoking in Belarus: evidence from a household survey. European Journal of Epidemiology 2001;17:245-53

131 Pudule I, Grinberga D, Kadziauskiene K, Abaravicius A, Vaask S, Robertson A et al. Patterns of smoking in the Baltic Republics. Journal of Epidemiology & Community Health 1999;53:277-82.

132 Kasmel A, Lipand A, Markina A. Health behaviour among Estonian adult population, Spring 2002. Estonian Centre for Health Education and Promotion, Tallinn, 2003.

133 Plieskiene A, Milasauskiene z, Klumbiene J. Relationship between smoking habits and other risk factors in Lithuanian rural population. Submitted to 3rd European Conference on Tobacco or Health, Warsaw 2002.

134 King A, Wold B, Tudor-Smith C & Harel Y. The Health of Youth: A cross- national survey. WHO Regional Publications, European Series No. 69, 1996. http://www.hbsc.org/downloads/HealthofYouth1.pdf

135 Hibell, B, Andersson B, Ahlstrom S, Balakireva O, Bjarnasson T, Kokkevi A, Morgan M. The 1999 ESPAD Report. Alcohol and other drug use among students in 30 European countries. Stockholm: Modin Tryck AB.

136 World Health Organization, Global Youth Tobacco Survey Fact Sheets. http://www.cdc.gov/tobacco/global/gyts/GYTS_factsheets.htm (last accessed 16/5/03)

137 Hearn MD, Prokhorov AV, Murray DM, Alexandrov AA, Luepker RV. Comparison of smoking prevalence in school students sampled from the United States of America and the Union of Soviet Socialist Republics. Int.J.Epidemiol. 1991;20:413-5

138 Currie C, Hurrelmann K, Settertobulte W, Smith R & Todd J. (eds). Health and health behaviour among young people. WHO Policy Series: Health policy for children and adolescents Issue 1. Copenhagen: WHO Regional Office for Europe, 2001. http://www.hbsc.org/downloads/Int_Report_00.pdf

347

139 Warren CW, Riley L, Asma S, Eriksen MP, Green L, Blanton C, Loo C, Batchelor S, Yach D. Tobacco use by youth: a surveillance report from the Global Youth Tobacco Survey project. Bulletin of the World Health ORganisation, 2000; 78:868-76.

140 Russia. In: Tobacco Health: a Global Status Report. Geneva: WHO, 1997. http://www.cdc.gov/tobacco/who/russianf.htm (last accessed 22/4/03)

141 Prokhorov AV, Alexandrov AA. Tobacco smoking in Moscow school students. Br.J.Addict. 1992;87:1469-76

142 Alexandrov AA, Alexandrova V, Ivanova EI. Prevalence of smoking among Moscow school children and approaches to prevention. I: Lu R, Mackay J, Niu S and Peto R (Eds). Tobacco: the growing epidemic. Procedings of the Tenth World Conference on Tobacco or Health, 24-28 August 1997, Beijing, China. London: Springer-Verlag, 266-8.

143 Levshin V, Drojachih V, Fedichkina T, Slepchenko N. Epidemioloy of smoking and some determinants of smoking behaviour. In: Lu R, Mackay, J, Niu S and Peto R (eds) Tobacco: the growing epidemic. Proceedings of the Tenth World Conference on Tobacco or Health, 24-28 August 1997, Beijing,China. London: Sringer-Verlag, 2000:52-4.

144 Hanson KP, Barchuk AS, Zabezhinski MA. Smoking among medical students and student nurses in the Russian Federation: Educational problems. In: Tobacco: the growing epidemic. Proceedings of the Tenth World Conference on Tobacco or Health, 24-28 August 1997, Beijing, China. Lu Mackay, Niu, Peto (eds). London: Springer, 2000, pp 923-4.

145 Peto R, Lopez A, Boreham J, Thun M, Clark HJ. Mortality from Smoking in Developed Countries 1950-2000: Indirect estimates from national vital statistics. Oxford: Oxford University Press, 1994.

146 Doll, R., Peto, R., Wheatley, K, Gray, R, and Sutherland, I. Mortality in relation to smoking: 40 years' observations on male British Doctors. BMJ 1994; 309: 901-911.

348

147 Peto R, Lopez AD, Boreham J, Thun M, Heath C. Mortality from smoking in developed countries: indirect estimation from national vital statistics. Lancet 1992;339;1268-78.

148 Murray C, Lopez A, eds. The Global Burden of Disease. Cambridge: Harvard School of Public Health, 1996.

149 Boyle P. Cancer, cigarette smoking and premature death in Europe: a review including the Recommendations of European Cancer Experts Consensus Meeting, Helsinki, October 1996. Lung Cancer 1997;17:1-60.

150 Wynder E,Graham E. Tobacco smoking as a possible etiologic factor in bronchiogenic carcinoma. JAMA 1950;143:329-38.

151 Levin, Goldstein, Gerhardt. Cancer and tobacco smoking. JAMA 1950;143:329- 36.

152 Mills, CA and Mills Porter, M. Tobacco smoking habits and cancer of the mouth and respiratory system. 10, 539-542. 1950.

153 Schrek, R, Baker, LA, Ballard, GP, and Dolgoff, S. Tobacco smoking as an etiological factor in disease. I. Cancer. Cancer Research 10, 49-58. 1950.

154 Doll R, Hill AB. Smoking and carcinoma of the lung. Preliminary report. 1950 [classical article]. Bulletin of the World Health Organization 1999;77:84-93.

155 Doll R, Hill AB. The mortality of doctors in relation to their smoking habits: a preliminary report. BMJ 1954;1:1451-55

156 Doll R, Hill AB. Lung cancer and other causes of death in relation to smoking. A second report on the mortality of British doctors. BMJ 1956;ii:1071

157 Hammond E, Horn D. The relationship between human smoking habits and death rates. JAMA 1954;155:1316-28.

158 Hammond E, Horn D. Smoking and death rates - report on forty-four months of follow-up of 187,783 men. I Total mortality. JAMA 1958;166:1159-72.

349

159 Hammond E, Horn D. Smoking and death rates - report on forty-four months of follow-up of 187,783 men. II death rates by cause. JAMA 1958;166:1294-308.

160 Doll R. Risk from tobacco and potentials for health gain. International Journal of Tuberculosis & Lung Disease 1999;3:90-9.

161 US Department of Health and Human Services. Reducing the health consequences of smoking: 25 years of progress. A report of the Surgeon General. US Department of Health and Human Services, Public Health Service, Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Health, DHSS Publication No. (CDC) 89-8411, 1989.

162 Thun MJ, Day-Lally CA. Calle EE, Flanders WD, Heath CW. Excess mortality among cigarette smokers: Changes in a 20-year interval. American Journal of Public Health 1995; 85:1223-30.

163 Stewart BW, Kleihues P (Eds). The World Cancer Report. World Health Organization International Agency for Research on Cancer. Lyon: IARC Press, 2003.

164 Wald NJ,.Hackshaw AK. Cigarette smoking: an epidemiological overview, Br.Med.Bull. 1996; 52:3-11.

165 Doll R. Cancers weakly related to smoking. Br.Med.Bull. 1996;52:35-49.

166 Lissowska J. Tobacco, alcohol, and diet in cancer etiology in Poland: a review of Polish epidemiological studies. Cancer Detect.Prev. 1993;17:485-94.

167 Peto R. Influence of dose and duration of smoking on lung cancer rates. In: IARC. Tobacco: a major international health hazard. IARC Scientific Publications no. 74. Lyon: IARC, 1986.pp 23-33.

168 Stellman SD. Cigarette yield and cancer risk: evidence from case control and prospective studies. In: IARC. Tobacco: a major international health hazard. IARC Scientific Publications no. 74. Lyon: IARC, 1986, pp197-209.

169 Peto R, Darby S, Deo H, Silcocks P, Whitley E, Doll R. Smoking, smoking cessation, and lung cancer in the UK since 1950: combination of national statistics with two case-control studies. BMJ 2000;321:323-9.

350

170 Harris JE, Thun MJ, Mondul AM, Calle EE. Cigarette tar yields in relation to mortality from lunc cancer in the cancer prevention study II prospective cohort, 1982-8. British Medical Journal 2004; 328: 72-6.

171 Townsend JL. Meade TW. Ischaemic heart disease mortality risks for smokers and non-smokers. J Epidemiol Community Health, 1979, 33: 243-7.

172 Lopez AD, Collishaw NE, and Piha T. A descriptive model of the cigarette epidemic in developed countries. Tobacco Control 3, 242-247. 1994.

173 US Department of Health and Human Services. The Health Benefits of Smoking Cessation: a report of the Surgeon General 1990. Maryland: CDC, 1990.

174 Gupta PC, Mehta HC. Cohort study of all-cause mortality among tobacco users in Mumbai, India. Bull World Health Organ 2000;78:877-83.

175 Chen ZM, Xu Z, Collins R, Li WX, Peto R. Early health effects of the emerging tobacco epidemic in China. A 16-year prospective study. JAMA 1997;278:1500-4.

176 Niu SR, Yang GH, Chen ZM, Wang JL, Wang GH, He XZ et al. Emerging tobacco hazards in China: 2. Early mortality results from a prospective study. BMJ 1998;317:1423-4.

177 Liu BQ, Peto R, Chen ZM, Boreham J, Wu YP, Li JY et al. Emerging tobacco hazards in China: 1. Retrospective proportional mortality study of one million deaths. BMJ 1998;317:1411-22.

178 Deev A, Shkolnikov V,. Neodnorodnost’ smertnosti: amaliz individualnikh dannyh. [Heterogeneity of mortality: analysis of individual-level data] In: Shkolnikov V., Andreev E., Maleva T. (Eds.) Neravenstvo i Smertnost v Rossii [Inequality and Mortality in Russia]. Moscow: Carnegie Center, 2000.

179 Britton A, McKee M. The relation between alcohol and cardiovascular disease in Eastern Europe: explaining the paradox. J Epidemiol Community Health 2000;54:328- 32.

180 Ezzati M, Lopez AD. Measuring the accumulated hazards of smoking: global and regional estimates for 2000. Tobacco Control 2003; 12: 79-85.

351

181 Cooper R, Schatzkin A. Recent trends in coronary risk factors in the USSR. American Journal of Public Health 1982; 72: 431-40.

182 Hense HW, Schulte H, Lowel H, Assmann G, Keil U. Framingham risk function overestimates risk of coronary heart disease in men and women from Germany--results from the MONICA Augsburg and the PROCAM cohorts. Eur Heart J 2003; 24: 937-45.

183 Brindle P, Emberson J, Lampe F, Walker M, Whincup P, Fahey T, Ebrahim S. Predictive accuracy of the Framingham coronary risk score in British men: prospective cohort study. BMJ 2003; 327: 1267.

184 Koek HLD, Bots ML, Grobbee DE. Are Russians different than other Europeans in their relation of risk factors to cardiovascular disease. European Journal of Epidemiology 2003; 18: 843-44.

185 Averina M, Nilssen O, Brenn T, Borx J, Kalinin AG, Arkhipovky LV. High cardiovascular mortality in Russia cannot be explained by the classical risk factors. The Arkhangelsk study 2000. Eur J Epidemiol 2003;18:871-8.

186 Pomerleau J, McKee M, Robertson A, Kadziauskiene K, Abaravicius A, Vaask S, Pudule I, Grinberga D. Macronutrient and food intake in the Baltic republics. Eur J Clin Nutr 2001: 55:200-7.

187 Zatonski WA, McMichael AJ, Powles JW. Ecological study of reasons for sharp decline in mortality from ischaemic heart disease in Poland since 1991. BMJ 1998;316:1047-51.

188 Bobak M, Skodova Z, Pisa Z, Poledne R, Marmot M. Political changes and trends in cardiovascular risk factors in the Czech Republic, 1985-92. J Epidemiol Comm Health 1997;51:272-7.

189 Bobak M, McKee M, Rose R, Marmot M. Alcohol consumption in a national sample of the Russian population. Addiction 1999;94:857-66.

190 Chenet L, Britton A, Kalediene R, Petrauskiene J. Daily variation in deaths in Lithuania: the possible contribution of binge drinking. Int J Epidmiol 2001;30:743-8.

352

191 Malyutina S, Bobak M, Kurilovitch S, Gafarov V, Simonova G, Nikitin Y, Marmot M. Relation between heavy and binge drinking and all-cause and cardiovascular mortality in Novosibirsk, Russia: a prospective cohort study. Lancet 2002;360:1448-54.

192 World Health Organization International Agency for Research on Cancer. Cancer incidence in the USSR (supplement to Cancer Incidence in Five Continents, Vol. 3). IARC Scientific Publications No. 48. Lyon: IARC, 1983.

193 Zaridze DG, Gurevicius R. Lung Cancer in the USSR: Patterns and trends. In: Zaridze DG, Peto R (eds). Tobacco: A major international health hazard. IARC Scientific publications NO. 74. Lyon: IARC, 1986.

194 La Vecchia C, Levi F. Lung cancer in Europe. European Journal of Public Health 2003; 13: 1-2.

195 Shkolnikov V, McKee M, Leon D, Chenet L. Why is the death rate from lung cancer falling in the Russian Federation? Eur J Epidemiol 1999; 15: 203-6.

196 Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2000: Cancer Incidence, Mortality and Prevalence Worldwide, Version 1.0. IARC CancerBase No. 5. Lyon, IARCPress, 2001. Available on: http://www-dep.iarc.fr/globocan/globocan.html (last accessed 9 December 2003)

197 Brennan P, Bray I. Recent trends and future directions for lung cancer mortality in Europe. British Journal of Cancer 2002; 87; 43-48.

198 Bobak M. Relative and absolute gender gap in all-cause mortality in Europe and the contribution of smoking. European Journal of Epidemiology 2003; 18: 15-18.

199 Cockerham WC, Snead MC, Dewaal DF. Health lifestyles in Russia and the Socialist Heritage. Journal of Health and Social Behavior 2002; 43: 42-55.

200 Andreev EM, Nolte E, Shkolnikov VM, Varavikova E, McKee M. The evolving pattern of avoidable mortality in Russia. Int J Epidemiol 2003; 32: 437-46.

201 Cooper R, Schatzkin A. The Pattern of mass disease in the USSR: A product of socialist of capitalist development. International Journal of Health Services 1982; 12: 459-80.

353

202 Puska P. Health promotion challenges for countries of the former Soviet Union: results from collaboration between Estonia, Russian Karelia and Finland. Health Promotion International 1995; 10: 219-228.

203 Paluosuo H. Health related lifestyles and alienation in Moscow and Helsinki. Social Science and Medicine 2000; 51: 1325-41.

204 O’Connor M. Tobacco control in countries of central and eastern Europe. In: Slama K (ed). Tobacco and Health. New York: Plenum Press, 1995, pp 273-7.

205 Barbour RS. The case for combining qualitative and quantitative approaches in health services research. J HealthServ Res Policy 1999; 4: 39-43.

206 Marshall C, Rossman GB. Designing qualitative research. 3rd Edition. London: Sage Publications, 1999.

207 Barbour RS. Mixing qualitative methods: quality assurance or qualitative quagmire. Qualitative health research 1998; 8: 352-61.

208 Gantley M. The qualitative and the quantitative: an anthropological perspective on research methodologies. Critical Public Health 1994: 5: 27-32.

209 United Nations Conference on Trade and Development (UNCTAD). World Investment Report 2001, Promoting Linkages. Geneva: United Nations, 2001.

210 http://cisid.who.dk/tobacco/ (last accessed 12 May 2003)

211 ERC Statistics International plc. The World Cigarettes I and II: The 2001 survey. Suffolk, Great Britain, 2001.

212 Volodina W, Alexeyev I. Meeting more than the market’s demand. Tobacco Journal International 2001; 3:27-35.

213 Volodina S. Ukraine: Mastering the crisis. Tobacco Journal International 2000: 4:64-70.

214 Anon. Filter cigarettes make headway on Ukraine’s turbulent market. World Tobacco for Russia and Eastern Europe 2000;1:3-7.

215 Anon. Tobacco traveller: Latvia. Tobacco Journal International 1994;3:66.

354

216 Anon. Tobacco traveller: Estonia. Tobacco Journal International 1994;3:64.

217 European Bank of Reconstruction and Development. Lithuania investment profile, 2001.

218 Anon. Kova plant closed. Tobacco Journal International 1996;4:58.

219 Glogan T. King Philip of Lithuania. Tobacco Journal International 2000; 1:40.

220 European Bank of Reconstruction and Development. Uzbekistan: investment profile, 2001.

221 Tuinstra T. Following the Silk Route. Tobacco Reporter 1998: October : 143-5.

222 Anon. Tobacco traveller: Uzbekistan. Tobacco Journal International 1996; 4:63.

223 United States Department of Agriculture, Foreign Agricultural Section. Uzbekistan, Republic of: Tobacco and Products Annual 2000. Gain Report #UZ0005. April 2000. (www.fas.usda.gov/gainfiles/200004/25667473.pdf)

224 Khan MA. New tobacco factory opened in Kazakhstan. Tobacco Journal International 2000;4:71-72.

225 Anon. Armenia: Canadian involvement. Tobacco Journal International 1997;4:4

226 http://www.export.am/Guides/Country_Prof/MainSect/FoodProc.htm. (last accessed 27/2/02)

227 European Bank of Reconstruction and Development. Armenia: Investment Profile, 2001.

228 http://www.agbu.org/pages/canada/canarmbus.html (last accessed 27/2/02)

229 Wright R. Cigarettes help build impoverished Armenia. The Russia Journal April 6-12, 2001. http://www.Russiajournal.com/weekly/pdfs/106_16.pdf (last accessed 4/4/02)

230 Anon. Azerbaijan: RJR in Joint venture. Tobacco Journal International 1996;4:4.

355

231 United States Department of Agriculture, Foreign Agricultural Section. Azerbaijan, Republic of: Tobacco and Products Annual Report 2000. GAIN Report #AJ0002.June 2000. (http://www.fas.usda.gov/gainfiles/200003/25637244.pdf)

232 Reuters. Investors reopen Azerbaijan’s sole tobacco plant. 3 February 2000. http://lists.essential.org/pipermail/intl-tobacco/2000q1/000038.html (last accessed 1 June 2003)

233 European Tobacco-Baku now country’s biggest taxpayer http://www.internationalspecialreports.com/ciscentralasia/01/azerbaijan/europeantobacc o.html (last accessed 3 June 2003)

234 Anon. Tobacco Traveller: Belorussia. Tobacco Journal International 1996; 4:61.

235 Hidalgo L. Turkmen face smoking ban. BBC News. (http://news.bbc.co.uk/hi/english/world/asia-pacific/newsid_602000/602581.stm)

236 Anon. Moldova: New bids considered. Tobacco Journal International 1997;2:6.

237 European Bank of Reconstruction and Development. Moldova: Investment profile. 2001.

238 Anon. Moldova Joint Venture. Tobacco Journal International 1996; 5:7.

239 Anon. Tobacco Traveller: Tajikistan. Tobacco Journal International 1996; 6:82.

240 European Bank of Reconstruction and Development. Kyrgyzstan: investment profile, 2001.

241 European Bank of Reconstruction and Development. Estonia: investment profile, 2001.

242 Connolly GN. Tobacco, trade and eastern Europe. In: Tobacco and Health. K Slama (ed). New York: Plenum Press, 1995.

243 Horne B. Russia strives to create more attractive climate for foreign investment. Tobacco Journal International 1995;6:12-18

244 Crescenti MG. A market for the brave. Tobacco Journal International 1997;6:14-21.

356

245 Tuinstra T. Light in the darkness. Tobacco Reporter 1999; August: 32-34.

246 Anon. Russia, BAT investment. Tobacco Journal International 1997;4:98.

247 Anon. 1998 revisited. Tobacco Journal International 1999; Yearbook: 4-8.

248 Anon. Rothmans’ new plant. Tobacco Journal International 1997; 6:8.

249 Volodina S. Showtime for tobacco. Tobacco Journal International 2000;5:90-92.

250 Anon. 1999 revisited. Tobacco Journal International 2000;Yearbook:6.

251 Anon. Major investments indicate continuing faith in local production. World Tobacco for Russia and Eastern Europe 1998;II:39-40.

252 Crescenti MG. Confidence in a Turbulent market. Tobacco Journal International 2000;1:26-27.

253 Hundley L. Investing. Tobacco Reporter 1995;September: 60-62.

254 Anon. Four multinationals, J Reynolds, Philip Morris, BAT and Reemtsma have purchased stakes in Ukrainian factories. Tobacco Reporter 1996; July:45-48.

255 Foreign involvement in the tobacco industries of Eastern Europe and the Former Soviet Union. (status August 1994). Tobacco Journal International 1994;5:97-104 (insert).

256 Phillip Morris International (http://www.pmintl.com/about_us/history.html last accessed 7/3/01)

282 Anon. 1996 revisited (Ukraine). Tobacco Journal International 1997; Yearbook: 8.

258 Horne B. Opposition to reform hinders economic growth in Ukraine. Tobacco Journal International 1995; 6:20-22.

259 TR Staff. Multinational investment . Cigarette company acquisitions in Central and Eastern Europe. Tobacco Reporter, 1998; Winter (bonus issue): 36-37.

260 Shchur, Vasyl. Foreign investment igniting tobacco market. Kyiv Post 23rd May 2002. (http://www.kpnews.com/displaypr.php?arid=11122, last accessed 12/9/02)

357

261 Anon. Late News: Lithuania new factory. Tobacco Journal International 1994;5: 112.

262 Glogan T. Baltic Tobacco industries struggle to develop. Tobacco Journal International 1995;4:54-58.

263 Anon. Tobacco Traveller: Kazakhstan. Tobacco Journal International 1996;6:81.

264 Horne B. Involvement in Eastern Europe and the FSU. Tobacco Journal International 1994;5:88-96.

265 Horne B. BAT forging ahead with factory in Samarkand. Tobacco Journal International 1997;1:20-22

266 Anon. Kyrygzstan. Tobacco Journal International 2001; Yearbook 2001: 8.

267 Mitreva M. Russia’s revival. Tobacco Reporter 2000; July: 48-51.

268 Crescenti MG. Russia Getting back on its feet. Tobacco Journal International 1999; 6:54-6.

269 http://www.corporateinformation.com/uasector/Advertising.html (last accessed 2nd January 2001)

270 http://www.adageglobal.com/cgi-bin/pages.pl?link=445(last accessed 5th April 2002)

271 Zenith Media. Central and Eastern Europe market and mediafact. London: Zenith Optimedia, 2000.

272 Zenith Media. Central and Eastern Europe market and mediafact. London: Zenith Optimedia, 2003.

273 Partsch C. New law bans TV ads for cigarettes, alcohol. St Petersburg Times, June 27-July 3 1995. (http://www.sptimesrussia.com/secur1/112/new-law.html) last accessed 29/1/01

274 World Tobacco File 1997. London: Market Tracking International Limited, 1999.

358

275 Kutuzov R. Rybak S, Neimysheva N. Duma dilutes anti-tobacco law. St Petersburg Times, June 19, 2001. (http://www.sptimesrussia.com/secur/679/news/b_3717.htm?725curr last accessed 29/11/01)

276 United States Department of Agriculture, Foreign Agricultural Service. Import requirements and restrictions for tobacco products in foreign markets. Supplement to Tobacco: world markets and trade. July 1999. (Available on http://www.fas.usda.gov/tobacco/circular/1998/impreqmts/imprqmts.htm).

277 WHO Regional Office for Europe. The European Report on Tobacco Control Policy. Review of implementation of the Third Action Plan for a Tobacco-free Europe 1997-2001. Copenhagen: WHO, 2002.

278 Stanikas T. Tobacco control in Lithuania. Undated. (http://health21.hungrary.globalink.org/filternonline/exvclusive/lithuania.html. last accessed 12/11/01).

279 Anon. Lithuania health warnings. Tobacco Journal International 1996;5:7.

280 World Bank. Economics of tobacco for the Europe and Central Asia (ECA) Region, May 2001. http://www1.worldbank.org/tobacco/pdf/country%20briefs/Eastern%20and%20Central %20Asia.pdf (last accessed 11/6/02)

281 Glogan T. Contraband impairs Estonian market. Tobacco Journal International 1999;5:69.

282 Glogan T. Latvia suffers under barrage of contraband. Tobacco Journal International 1999;6:95-6.

283 Talov I. Portraits of producers. Business, 26 February 2001 [Russian]. http://www.business.ua/i424/a8273/ (last accessed 9 July 2003)

284 Talov I. Industrial trade survey of the tobacco industry. Business, 26 February 2001 [Russian]. www.business.ua/i424/a8272 (last accessed 9 July 2003)

359

285 Anon. Statistics: production of cigarettes. Tobacco Journal International 1997;6:87-90.

286 United States Department of Agriculture, Foreign Agricultural Service. World Tobacco Markets Circular series FT-3-01 Sept 2001.

287 British American Tobacco. Quarterly Report to 30 September 1998. London: British American Tobacco, 1998.

288 British American Tobacco. Quarterly Report to 31 March 1999. London: British American Tobacco, 1999.

289 British American Tobacco. Quarterly Report to 30 September 2000. London: British American Tobacco, 2000.

290 British American Tobacco. Quarterly Report to 31 March 2001. London: British American Tobacco, 2001.

291 Philip Morris. Excerpts from the Philip Morris Companies Inc. 2000 Fact Book. (http://www.pmint.com/fin_overview/book.html last accessed 6/11/01)

292 Edgecliffe-Johnson A. Discounting fails to dent Philip Morris. Financial Times October 18th 2001 p 31.

293 Parrott B. Perspectives on postcommunist democratisation. In: Conflict, cleavage and change in Central Asia and the Caucasus. Dawisha K, Parrott B (eds). Cambridge: Cambridge University Press, 1997.

294 Uzbekistan: who’s in charge now? Tobacco Control 2000;9:359-6.

295 Anon. Health and dictatorship: effects of repression in Turkmenistan . Lancet 2003; 361: 69-70.

296 A tactical market. Interview with Denis Gourinovich, corporate relations manager at BAT in Belarus. Tobacco Journal International 12/2/03 http://www.tobaccojournal.com/show_artikel.php3?id=2971 (last accessed 6 June 2003)

360

297 Chossudovsky M. The “Thirdworldisation” of the Russian Federation. In: The globalisation of poverty, impacts of IMF and World Bank Reforms. London: Zed books, 1998.

298 Council on Scientific Affairs. The worldwide smoking epidemic: tobacco trade, use and control. JAMA 1990;263:3312-18. {401}

299 Novotny T, Zhao F. Consumption and production waste: another externality of tobacco use. Tob Control 1999;8:75-80

300 http://apps.fao.org/page/collections?subset=agriculture. (last accessed 10th July 2001)

301 United States Department of Agriculture, Foreign Agricultural Service. http://usda.mannlib.cornell.edu/data-sets/specialty/94012/ (last accessed 13/1/02)

302 http://apps.fao.org/page/form?collection=Production.Crops.Primary&Domain=Pr oduction&servlet=1&language=EN&hostname=apps.fao.org&version=default (last accessed 10th July 2001)

303 http://apps.fao.org/page/form?collection=Trade.CropsLivestockProducts&Domai n=Trade&servlet=1&language=EN&hostname=apps.fao.org&version=default (last accessed 16 September 2002)

304 Department of International Economic and Social Affairs, Statistical Office. Demographic Yearbook historical supplement. New York: United Nations, 1979.

305 Department of International economic and Social Affairs, Statistical Office. 1987 Demographic Yearbook. New York: United Nations, 1989.

306 Department of International economic and Social Affairs, Statistical Office. 1990 Demographic Yearbook. New York: United Nations, 1992.

307 Anon. East European production can be revived. World Tobacco for Russia and Eastern Europe 1996;II:10-16.

361

308 Prislopeanu I. Eastern Europe: Farewell to monopolies. Tobacco Journal International 2000:5:7-13.

309 Shkolnikov V, Nemtsov A. The anti-alcohol campaign and variations in Russian mortality. In: Bobadilla JL, Costello CA, Mitchell F (eds). Premature death in the New Independent States. Washington DC: National Academy Press, 1997.

310 USDA FAS. Azerbaijan, Republic of: Tobacco and Products Annual Report 2000. GAIN Report #AJ0002.June 2000. (http://www.fas.usda.gov/gainfiles/200003/25637244.pdf)

311 Anon. Move towards local sourcing. Tobacco Journal International 1999;4:81.

312 Anon. Leaf output growing in Uzbekistan. Tobacco journal international 1999;5:28.

313 Statistical Committee of the Commonwealth of Independent States. External trade of the CIS countries. Moscow: Mezhgosudarstvennyĭ statisticheskiĭ.

314 Joossens L, Raw M. Cigarette smuggling in Europe: who really benefits? Tobacco Control 1998; 7: 66-71.

315 United States Department of Agriculture, Foreign Agricultural Service. Russian Federation Tobacco and Products Annual 2002. Gain Report #RS2014, 2002.

316 World Tobacco Trends. Henley on Thames: NTC Publications Limited, 1998.

317 International Trade Publications Ltd. World Tobacco File 1998-2001. London: DMG Business Media, 1998.

318 World Health Organisation. Russia. In: Tobacco Health: a Global Status Report. Geneva: WHO, 1997.

319 British American Tobacco. Moldova marketing plan 1994-2003. Discussion Draft. [01/03/1994]. Guildford Depository. BAT. Bates No: 602750435-51

320 Herter U, Brookes N. BAT Industries – Uzbekistan. British American Tobacco. 29th November 1993. Bates No.: 202208940 -2.

362

321 British American Tobacco. Draft 1. Project Umpire skeleton business plan. British American Tobacco. [1994] Bates no.: 301734376-4406

322 Marr M, Lewis L. Russia Marketing/distribution study: Yava Factory, Nov. 23 – Dec. 6 1992. 25/01/1993. Guildford Depository. BAT. Bates No: 500229032-98.

323 British American Tobacco. Russia Business outlook (facts and forecasts). [June 1994]. Guildford Depository. BAT. Bates No: 203466112-44.

324 Bain Link & Co.. BAT Industries Russian distribution strategy final presentation. 21/09/1993. Guildford Depository. BAT. Bates No: 301755203-385.

325 Anon. Russia’s Leaf imports rise as cigarette sale fall. World tobacco for Russia and Eastern Europe 1991;1:39-46.

326 Garlick P. Tobacco for the Russian market. Tobacco Journal International 2000;3:34-36.

327 The World Bank. Curbing the Epidemic: Governments and the Economics of Tobacco Control. Washington D.C: The World Bank, 1999,.

328 Murray CJL, Lopez AD. Mortality by cause for eight regions of the world: global burden of disease study. Lancet 1997; 359: 1269-76.

329 World Health Organisation Regional Office for Europe. Health For All Database. Copenhagen, 2003. http://hfadb.who.dk/hfa/ (last accessed 15 May 2003).

330 World Health Organisation. Tobacco or health: a global status report. Country profiles by region 1997. http://www.cdc.gov/tobacco/who/armenia.htm (last accessed 22 April 2003)

331 URL: http://www.llh.at/llh_partners_start.html (accessed 10 July 2003)

332 Post-communist barometer surveys. Glasgow: Centre for the Study of Public Policy, University of Strathclyde (http://www.cspp.strath.ac.uk).

333 Heatherton TF, Kozlowski LT, Frecker RC, Fagerstrom KO. The Fagerstrom Test for Nicotine Dependence: a revision of the Fagerstrom Tolerance Questionnaire.Br J Addict. 1991 Sep;86:1119-27.

363

334 The Fagerstrom test for nicotine dependence http://www.fpnotebook.com/PSY81.htm (last accessed 9/5/03)

335 Badurashvili I, McKee M, Tsuladze G, Meslé F, Vallin J, Shkolnikov V. Where there are no data: what has happened to life expectancy in Georgia since 1990? Public Health 2001; 115: 394-400.

336 Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The validity of self-reported smoking: a review and meta-analysis. Am J Public Health 1994; 84: 1086-93.

337 Laatikainen T, Vartiainen E, Puska P. Comparing smoking and smoking cessation process in the Republic of Karelia, Russia and North Karelia, Finland. J Epidemiol Community Health 1999;53:528-34.

338 World Tobacco file 1997 – cigars, smoking tobacco and smokeless tobacco. London: DMG Business Media, 1999.

339 Alcohol and Drug Information Center (ADIC- Ukraine). Economics of tobacco control in Ukraine from the public health perspective. Kiev: ADIC, 2002. http://www.adic.org.ua/adic/reports/econ/ch-3/3-3.htm (last accessed 17 April 2003).

340 Molarius A, Parsons RW, Dobson AJ, Evans A, Fortmann SP, Jamrozik K et al. Trends in cigarette smoking in 36 populations from the early 1980s to the mid-1990s: findings from the WHO MONICA Project. Am.J.Public Health 2001; 91:206-212.

341 Cockerham WC. Health lifestyles in Russia. Soc Sci Med 2000; 51:1313-1324.

342 Erens B, Primatesta P. Health Survey for England: Cardiovascular diseases 98. Volume 1. Findings. London: The Stationary Office, 1999.

343 Cavelaars AEJM, Kunst AE, Geurts JJM, Crialesi R, Grotvedt L,Helmert U, et al. Educational differences in smoking: international comparison. Br Med J 2000; 320: 1102-7.

344 Hadju P, McKee M, Bojan F. Changes in premature mortality differentials by marital status in Hungary and in England and Wales. Eur J Publ Health 1995; 5:259-264.

364

345 Watson P. Explaining rising mortality among men in Eastern Europe. Soc Sci Med 1995; 41:923-934.

346 Pakriev S, Shlik J. Unpublished report – Patterns of tobacco use in the rural Udmurtia, 2001.

347 Palosuo H., Uutela A., Zhuravleva I., Lakomova N. Social patterning of ill health in Helsinki and Moscow. Soc Sci Med 1998; 46:1121-1136.

348 Eng PM, Rimm EB, Fitzmaurice G, Kawachi I. Social ties and change in social ties in relation to subsequent total and cause-specific mortality and coronary health disease incidence in men. Am J Epidemiol 2002; 155:700-709.

349 Hu Y, Goldman N. Mortality differentials by marital status: an international comparison. Demography 1990; 27:233-250.

350 ERC Group plc. World Cigarettes 2001 (volume 1). Newmarket: ERC Group, 2001.

351 Anon. A tactical market. Tobacco Journal International. 12/2/2002. http://www.tobaccojournal.com/show_artikel.php3?id=2971 (last accessed 15 July 2003)

352 Chantornvong S, McCargo D. Political economy of tobacco control in Thailand. Tobacco Control 2001; 10: 48-54.

353 Glantz SA, Barnes DE, Bero L, Hanauer P and Slade J. Looking through a keyhole at the tobacco industry: The Brown and Williamson documents, JAMA 1995;274: 219-24.

354 Rabin RL, The third wave of tobacco tort litigation. In: Rabin RL, Sugarman SD (eds). Regulating Tobacco. New York: Oxford University Press, 2001.

355 The Tobacco Control Resource Centre, Inc at Northeastern University School of Law. Kelder G, Davidson P (eds). The Multistate Master Settlement Agreement and the future of state and local tobacco control: an analysis of selected topics and provisions of the Multistate Master Settlement Agreement of November 23, 1998. 1999 http://www.tobacco.neu.edu/msa/msa_analysis.pdf (last accessed 31 October 2003)

365

356 MacKenzie R, Collin J and Lee K (2003) The tobacco industry documents: an introductory handbook and resource guide for researchers. London: London School of Hygiene & Tropical Medicine. http://www.lshtm.ac.uk/cgch/tobacco/Handbook%2008.07.03.pdf (last accessed 31 October 2003)

357 Malone RE and Balbach ED. Tobacco industry documents: treasure trove or quagmire? Tob Control 2000;9: 334-38 http://tc.bmjjounals.com/cgi/content/full/9/3/334

358 Office of the Attorney General State of California Master Settlement Site http://caag.state.ca.us/tobacco/msa.htm

359 Balbach ED, Gasior RJ and Barbeau EM. Tobacco industry documents: comparing the Minnesota Depository and internet access. Tob Control 2002;11: 68-72.

360 Lee K, Gilmore A, Collin J. Research of the Guildford Depository: Revealing insights from the Guildford Depository. 2004; 99:394-7.

361 Muggli M, LeGresley E, Hurt R. Big Tobacco is Watching: British American Tobacco’s Surveillance and Information Concealment at the Guildford Depository. The Lancet 2004 363:1746-7.

362 Collin J, Lee K, Gilmore A. Unlocking the corporate documents of British American Tobacco: An invaluable global resource needs radically improved access. Lancet 2004; 363: 1746-7.

363 Drew D. Lovells. Depository Report No 250. February 29, 2000. British American Tobacco. Bates No.:321669531- 33. Minnesota Tobacco Document Depository.

364 http://www.channel4.com/news/2004/10/week_2/04_tobacco.html

365 Birnbauer W. Dirty play by tobacco giant, says insider. Sydney Morning Herald July 19, 2003.

366 Yallop R, Haslem B. More heat on tobacco firms The Australian July 23, 2003

366

367 Meltzer, J. Letter from Ogilvy Renault: 5th June 1992. Lovell White Durrant. Guildford Depository. Bates no. 20313418-21. 5 June 1992.

368 Select Committee on Health. The Tobacco industry and the health risks of smoking: second report. June, 2000: http://www.parliament.the-stationery- office.co.uk/pa/cm199900/cmselect/cmhealth/27/2702.htm (accessed 28 April 2005)

369 http://www.lshtm.ac.uk/cgch/tobacco/guildford.htm

370 http://bat.library.ucsf.edu/ (last accessed 18 April 2005)

371 WHO, Tobacco Industry Strategies to undermine tobacco control activities in the World Health Organization. Tobacco Free Initiative, 2000 http://www5.who.int/tobacco/page.cfm?sid=58

372 Givel M, Glantz S. The ‘Global Settlement” With the Tobacco Industry: 6 Years Later. American Journal of Public Health 2004; 94(2): 218-24. 373 World Health Organization. Towards Health with Justice: Litigation and public inquiries as tools for tobacco control. Geneva: WHO;2002 374 Bero L. Implications of the tobacco industry documents for public health and policy. Annu Rev Public Health 2003; 24: 267-88.

375 O’Sullivan B, Chapman S. Eyes of the prize: transnational tobacco companies in China 1976-1997.Tob. Control, Sep 2000; 9: 292 - 302.

376 Barnoya J, Glantz S. Tobacco industry success in preventing regulation of secondhand smoke in Latin America: the "Latin Project". Tob Control 2002; 11: 305-14.

377 Hiilamo H. Tobacco industry strategy to undermine tobacco control in Finland. Tob Control 2003; 12: 414-23.

378 Szilagyi T. Strategies and tactics of transnational tobacco companies in Hungary, 1990-2001. MPhil thesis. Sydney: University of Sydney, 2003.

379 Lee C-Y, Glantz S. The Tobacco Industry's Successful Efforts to Control Tobacco Policy Making in Switzerland. San Francisco CA: Center for Tobacco Control Research and Education, 2001. http://repositories.cdlib.org/ctcre/tcpmi/Swiss2001

367

380 Committee of Experts. Tobacco Company Strategies to Undermine Tobacco Control Activities at the World Health Organization. Geneva: WHO, July 2000.

381 British American Tobacco. Annual Review and summary financial statement 2001. London: BAT, 2002. (http://www.bat.com/AnnualReport2001/annrev/13_download/Review.pdf last accessed 27/02/03)

382 Glantz, S. The truth about big tobacco in its own words. BMJ 2000;316: 1553 http://www.bmj.com/cgi/reprint/321/7257/313.pdf

383 Campbell D. BAT Industries Plc in the 1980s and 1990s. In: Stonehouse G, Hamill J, Campbell D and Purdie T Global and Transnational Business: Strategy and Management. Chichester: Wiley, 2000.

384 Hill MR. Archival strategies and techniques. Qualitative research methods volume 31. Newbury Park, California: Sage Publications, 1993.

385 Forster N, The Analysis of Company Documentation. In: Cassell C, Symon G (eds). Qualitative Methods in Organizational Research. A practical guide. London: Sage Publications, 1994.

386 Foster J. Report on British American Tobacco Guildford Depository. London: London School of Hygiene and Tropical Medicine, 2000.

387 Levin M. Archive a Smoking Gun for Tobacco Firm. Why British American released papers showing possible links to contraband sales remains a mystery. LA Times November 23, 2003

388 Dey I. Chapter 14: Corroborating evidence. In: Qualitative data analysis: A user- friendly guide for social scientists. London: Routledge, 1993.

389 Marx K, Engels F. Communist Manifesto. London: Lawrence and Wishard, 1983.

390 European Bureau for Action on Smoking Prevention. Test the East. The tobacco industry and Eastern Europe. Brussels: BASP, 1991

368

391 “Emerging Markets” In: Tobacco Explained – the truth about the tobacco industry in its own words. London: Action on Smoking and Health, 1998.

392 British American Tobacco. BAT Bulletin November 1994. Guildford Depository. BAT. Bates No: 500316187-8.

393 British American Tobacco. Slide material for DSW presentation 17.2.92: Royal Berkshire Hotel. 17/2/92. Guildford Depository. BAT. Bates No: 203471856 - 868.

394 British American Tobacco. BATCO Operating Group Five Year Plan 1993-1997. [1993]. Guildford Depository. BAT. Bates No: 203481292-365.

395 Van Waay A. Ukraine - Marketing. 23/04/1992. Guildford Depository. BAT. Bates No: 301689859-61.

396 Department of Trade and Industry. Joint Ventures in Eastern Europe. August 1989. Guildford Depository. BAT. Bates No: 503078330-47.

397 KPMG Management Consulting. A study of the Russian Privatisation Process: Changing enterprise behaviour. BAT. 1993. Bates No: 203465968-203466040.

398 Lowe, Roger. [Memo to BAT re KPMG study on Russian privatisation] BAT. 27 May 1993. Bates no.: 203465967.

399 British American Tobacco. Study on Eastern Europe. 18/01/1990. Guildford Depository. BAT. Bates No: 203825453-92.

400 INFOTAB. Eastern Europe: Lifting the Curtain. October 1990. Guildford Depository. BAT. Bates No: 203825494-509.

401 Coplin WE, O'Leary MK (eds). East Europe & the Republics. A Political Risk Annual. First Edition. New York: Political Risk Services, 1992. Guildford Depository. BAT. Bates No: 201789689-704.

402 Marr M. Kyrgyzstan: Domestic market forecasts by segment. 30/08/1994. Guildford Depository. BAT. Bates No: 602750430-2.

369

403 Marr M. Uzbekistan. Preliminary marketing report following visit to Tashkent & Samarkand 26-30 July. 11/08/1993. Guildford Depository. BAT. Bates No: 203832508- 16.

404 Barnes M. Market research project - Executive summary - phase 2 (quantitative). Evaluation of Java, Cosmos and other domestic brands in Russia. 29/07/1993. Guildford Depository. BAT. Bates No: 500228841-53.

405 Bielefeldt R. BAT marketing strategy for the Russian Federation. [1994]. Guildford Depository. BAT. Bates No: 500229227-41.

406 Moscow Sociological Agency. Special Group Discussions - Uzbekistan. Summary report. [06/09/1994]. Guildford Depository. BAT. Bates No: 203457726-39.

407 British American Tobacco. New tobacco markets. 13/12/1991. Guildford Depository. BAT. Bates No: 203469386-94.

408 Van Waay A. Ukraine: Marketing strategy. 05/05/1992. Guildford Depository. BAT. Bates No: 301689856-8.

409 J Henry Schroder Wagg & Co. Ltd. Preliminary review of the Transcaucasian and Central Asian Republics of the former Soviet Union. Volume I. August 1992. Guildford Depository. BAT. Bates No: 203456565-86.

410 British American Tobacco. Belarus: Note of a meeting held on 6 March 1995 at Windsor House. 20/03/1995. Guildford Depository. BAT. Bates No: 203477045-7.

411 British American Tobacco. Future leaf requirements of BAT operations in the CIS. Guildford Depository. BAT. Bates No: 602750626-8.

412 Barton H. NBD Strategy. 18/11/1992. Guildford Depository. BAT. Bates No: 201789654-8.

413 Haslam F. Note for TEC. Potential involvement with USSR. 20/02/1989. Guildford Depository. BAT. Bates No: 203827396-411.

414 Haslam F. Potential involvement with the USSR. 17/6/1991. Guildford Depository. BAT. Bates No: 203827394-5.

370

415 Barton H. Draft BAT Industries Tobacco Strategy (Tobacco Strategy Group 19th May 1993, Agenda item 7. 12/05/1993. Guildford Depository. BAT. Bates No: 203469465-85.

416 British American Tobacco. Eastern Europe/USSR. 27/09/1991. Guildford Depository. BAT. Bates No: 201789795-7.

417 Tomat H. Visit to Moscow from 26th to 30th November 1990. Visit to Kiev from 10th to 11th December 1990. 20/12/1990. Guildford Depository. BAT. Bates No: 203827452-63.

418 Brookes N. Note to the CEC. Proposed BAT Tobacco investment strategy in Russia. [1993]. Guildford Depository. BAT. Bates No: 203466243-55.

419 Stevens B. Secret. Minutes of the meeting on Eastern Europe 9.00am on 23.1.91, Westminster House, London. 24/01/1991. Guildford Depository. BAT. Bates No: 203827437-41.

420 Unknown . CIS Recommendation BATCo Brands. [1993]. Guildford Depository. BAT. Bates No: 500209752-95.

421 British American Tobacco. New Tobacco markets. 3/6/1992. Guildford Depository. BAT. Bates No: 203469395-403.

422 Van Waay A. Brand strategies E. Europe and former Soviet Union. Guildford Depository. BAT. 6 December 1991. Bates No. 203471338.

423 Marr M. Moldova marketing plan 1994-2003. Discussion Draft. March 1994. Guildford Depository. BAT. Bates No: 602750435-51.

424 Direct Marketing Company Group. Press and TV Monitoring. Report on TV monitoring. Report for the month of January. [1993]. Guildford Depository. BAT. Bates No: 500229012-27.

425 Phillips J. Club 555. 18/05/1994. Guildford Depository. BAT. Bates No: 500229110-7.

426 Titmuss V. CIS- Road to Wembley English FA Cup Sponsorship - 555. 14/01/1993. Guildford Depository. BAT. Bates No: 500209748-9.

371

427 Tomat H. Public Affairs in the CIS. 05/05/1993. Guildford Depository. BAT. Bates No: 502550121-2.

428 Marr M. Uzbekistan - BAT Group brand presence. 04/02/1994. Guildford Depository. BAT. Bates No: 203459242-3.

429 Barton H. New business development strategy. 11/05/1993. Guildford Depository. BAT. Bates No: 203469421-8.

430 Van Waay A. Secret. Note to Mr DS Watterton. Soviet Union. 19/02/1992. Guildford Depository. BAT. Bates No: 203471295-9.

431 Kholmogorova GT, Prokhorov AV. West goes East: the new tobacco situation in Russia. Tobacco Control 1994; 3: 145-7.

432 Tomat HA. Exports to Russia. 08/01/1991. Guildford Depository. BAT. Bates No: 203827435-6.

433 Zimmerman C. Update: Philip Morris. Pioneering the "New Europe". Tobacco Reporter February 1992. Guildford Depository. BAT. Bates No: 203827368-70.

434 Tomat H. Memo for discussion with Sir Patrick. Re: USSR. 15/01/1991. Guildford Depository. BAT. Bates No: 203827448-9.

435 van Waay A. Notes of meeting held at Windsor House, 11.2.92. EXA International, Paris. 12/02/1992. Guildford Depository. BAT. Bates No: 203827381.

436 http://www.europa.eu.int/comm/europeaid/projects/tacis/foreword_en.htm (last accessed 6/3/03)

437 Rowlands-Rees G. Uzbekistan visit notes - 6th to 17th December 1993. 20/12/1993. Guildford Depository. BAT. Bates No: 203835038-41.

438 Marr M. Meeting Notes: MJM/Chairman Uzbakalea 22.2.94 Tashkent. 02/03/1994. Guildford Depository. BAT. Bates No: 203459229-30.

439 Rawstorne P. BAT poised to light up new market. 4/11/92. Guildford Depository. BAT. Bates No: 201789688.

372

440 Herter U. Note to the TSG 1c./11.5.94. Central Asia and the Caucasus. 4/5/94. Guildford Depository. BAT. Bates No: 500036164-6.

441 Abery S. Central/Eastern Europe Briefing Note. 26/08/1992. Guildford Depository. BAT Bates No.: 502577451-6. [

442 The International Consortium of Investigative Journalists. Tobacco companies linked to criminal organisations in lucrative cigarette smuggling. 2001, http://www.publici.org/story_01_030301_txt.htm (last accessed 13/2/03)

443 Beelman MS, Campbell D, Ronderos MT, Schelzig EJ. Major tobacco multinational implicated in cigarette smuggling, tax evasion, documents show. International Consortium of Investigative Journalists, 2000. http://www.publicintegrity.org/dtaweb/report.asp?ReportID=80&L1=30&L2=10&L3=1 0&L4=0&L5=0&Display=ICIJ (last accessed 31/10/03)

444 Unknown author. “Re: Transit Study”. Guildford depository. BAT 25th August 1989. Bates No.: 302000021 – 2.

445 BAT (UK and Export Limited). Company Plan 1993-1997. October 1992. Guildford Depository. BAT. Bates No: 502573663-719.

446 Unknown. BAT 1994 Media Meeting and CIS Operating Plan 1994. 07/12/1993. Guildford Depository. BAT. Bates No: 500209663-87.

447 Bain Link & Co.. Russian Distribution Strategy. [1993]. Guildford Depository. BAT. Bates No: 301755132-42.

448 Abery S. Re: 1995 plan. 30/06/1994. Guildford Depository. BAT. Bates No: 500229222-4.

449 Smith S. Questions raised by feasibility study, investment proposal and audit reports on Uzbekistan. May 1994. BAT. Bates No: 503905969-72.

450 Beelman MS, Campbell D, Ronderos MT, Schelzig EJ. Global reach of tobacco company’s involvement in cigarette smuggling exposed in company papers. International Consortium of Investigative Journalists, 2000.

373

http://www.publicintegrity.org/dtaweb/report.asp?ReportID=81&L1=30&L2=10&L3=1 0&L4=0&L5=0&Display=ICIJ (last accessed 31/10/03)

451 Aitken J. [Telex re Sino-Russian border supply]. 19/11/86. Guildford Depository. BAT. Bates No: 301634284.

452 Pong W. General Managers Monthly Review. January 1989 China. 2/2/1989. Guildford Depository. BAT. Bates No: 301634045-8.

453 Pong W. China. March 1989. Guildford Depository. BAT. Bates No: 301634036-8.

454 Coburn R. Hong Kong visit - 27th August 1986. 27/8/86. Guildford Depository. BAT. Bates No: 301634332-3.

455 Coburn R. File Note; Hong Kong meeting - 22nd September 1986. 25/9/86. Guildford Depository. BAT. Bates No: 301634322-4.

456 Aitken J. [Letter re Sino Russian border trade]. 28/10/86. Guildford Depository. BAT. Bates No: 301634282-3.

457 Cheung C. Heilongyiang Trip. [1/9/86]. Guildford Depository. BAT. Bates No: 301634326-8.

458 Aitken DJ. Minutes of BAT Central Asia/BATCo/BATCF/B&W/NBD Meeting held in Tashkent 10th June 1994. 14/06/1994. Guildford Depository. BAT. Bates No: 203459365-74.

459 SIS International. RJR Nabisco Holdings Competitive Intelligence Profile Prepared for Phillip Morris (sic) USA. April 1993. Philip Morris. Access Date: 25 June 2002. Bates No.: 2045122238-60. URL: http://legacy.library.ucsf.edu/cgi/getdoc?tid=efg03e00&fmt=pdf&ref=results

460 RJ Reynolds. Former Soviet Union 1996 operating plan. [1995]. RJ Reynolds. Access date: 19 November 2003. Bates no.: 514870515-624. URL: http://legacy.library.ucsf.edu/tid/gtx03a00

461 Amos A, Haglund M. From social taboo to “torch of freedom”: the marketing of cigarettes to women. Tobacco Control 2000; 9:3-8.

374

462 Bialous SA. Advertising restrictions: getting them right can help business. Alliance Bulletin Issue 36. Geneva: Framework Convention Alliance, 17th February 2003. http://www.fctc.org/newsletterINB6.shtml (last accessed 15th April 2003)

463 Koriukin K. Cracking a new customs code. The Moscow Times December 3rd 2001.

464 Border bore. The Economist December 15th 2001.

465 Arvedlund E. Tax collection under fire. St Petersburg Times, July 22-28 1996. http://www.sptimesrussia.com/secur/181-182/tax.html?725curr (last accessed 29/11/01)

466 Korchagina V. Luzhkov appoints ex-sports fund boss. St Petersburg Times August 4-10 1997. http://www.sptimesrussia.com/secur/284-285/luzhkov- appoints.html?725curr (last accessed 29/11/01)

467 Taibbi M. Ousted chief of Sports Fund shot, stabbed. St Petersburg Times. Juen 23-30 1996. (http://www.sptimesrussia.com/secur/171-172/ousted.html?725curr (last accessed 19/11/01)

468 Gordeyev A. A blessed business. St Petersburg Times, November 18-24, 1996. http://www.sptimesrussia.com/secur/213-214/bc.html?725curr (last accessed 29/11/01)

469 Warren M. Criminals 'cashing in on Orthodox Church business empire'. Daily Telegraph, 30 Jun 2000. http://www.telegraph.co.uk/news/main.jhtml?xml=%2Fnews%2F2000%2F06%2F30%2 Fworth30.xml (last accessed 3 July 2003)

470 Group Planning Department. BAT Industries - Future Business Environment 1996. Secret. 06/04/1995. Guildford Depository. BAT. Bates No: 500000489-526.

471 Volodina W, Alexeyev I. Meeting more than the market’s demand. Tobacco Journal International 2000: 3:27-35.

472 Salter R. New tobacco markets. 13/12/1991. Guildford Depository. BAT. Bates No: 203469386-94.

375

473 Brookes N. Note to Chief Executives Committee: Organisation of new business development. 01/09/1993. Guildford Depository. BAT. Bates No: 203469314-24.

474 Barton H. New business development strategy. 17/05/1993. Guildford Depository. BAT. Bates No: 203469331-5.

475 Barton H. Tobacco Strategy Group. 11/12/1992. Guildford Depository. BAT. Bates No: 203469531-3.

476 British American Tobacco. 1992 Business Review. 1992. Guildford Depository. BAT. Bates no. 203481215 – 248.

477 Sims D. Memorandum. 17/10/1993. Guildford Depository. BAT. Bates No: 203831282-3.

478 British American Tobacco. Meeting at the premises of BP Exploration Company Ltd., London. 27/05/1992. Guildford Depository. BAT. Bates No: 203461425-6.

479 Barton H. New Business Development - Market prioritisation. 03/12/1992. Guildford Depository. BAT. Bates No: 201789687.

480 Herter U. Note to the board of BAT Industries plc: Central Asian republics – briefing paper. 27/01/1994. Guildford Depository. BAT. Bates No: 503801076-8.

481 Johnson A. [Letter re Grodno tobacco factory and BAT investment in Belarus]. 01/11/1995. Guildford Depository. BAT. Bates No: 203476963-6.

482 British American Tobacco. Protocol of the negotiations between the representatives of the Ministry of Agriculture and Food of Belarus, Belpischeprom, Tobacco Factory Gordno and BAT Industries plc. 12/10/1994. Guildford Depository. BAT. Bates No: 203476888.

483 Lioutyi A. Belarus. 30/11/1995. Guildford Depository. BAT. Bates No: 203476960-1.

484 Mihalisko KJ. Belarus: retreat to authoritarianism. In Dawisha K, Parrott B (eds.) Democratic changes and authoritarian reactions in Russia, Ukraine, Belarus and Moldova.

376

485 Hewko J. [memo re developments in Belarus]. 14/04/1995. Guildford Depository. BAT. Bates No: 203477007-8.

486 Van Waay A, Bruce-Miller N, Booker M. Secret. Report Uzbekistan 18th to 21st May 1993. 28/05/1993. Guildford Depository. BAT. Bates No: 203832342-65.

487 Lioutyi A. Note for head of Project Uzbekistan Jim Rowland -Rees. 20/12/1993. Guildford Depository. BAT. Bates No: 203835035-7.

488 British American Tobacco. New Business Development - 1992 Strategy Review. [1992]. Guildford Depository. BAT. Bates No: 201789662-8.

489 Harington G. Schroders' Visit to Kiev: 13th to 16th October 1992. 21/10/1992. Guildford Depository. BAT. Bates No: 203837193-8.

490 British American Tobacco. Draft note to CEC. Valuation of Tobacco Projects In Eastern Europe and the Former Soviet Union. 01/10/1993. Guildford Depository. BAT. Bates No: 202208976-83.

491 Brookes N. Note to Mr U G V Herter. 22/06/1994. Guildford Depository. BAT. Bates No: 600507684-5.

492 British American Tobacco. YAVA - Share acquisition proposal. [10/9/93]. Guildford Depository. BAT. Bates No: 203466222-8.

493 Watterton D. Secret. New business development progress report - April 1992. 14/04/1992. Guildford Depository. BAT. Bates No: 203471748-56.

494 Commersant Daily. Philip Morris to buy tobacco factory after all. 20/8/93. Guildford Depository. BAT. Bates No: 203466234-5.

495 Whitehouse A. BAT/Belarus Negotiating Parameters. Guildford Depository. BAT. Bates No: 203477144-8.

496 Herter U. Note to Chief Executive's Committee. Uzbekistan. 02/03/1994. Guildford Depository. BAT. Bates No: 503801062-73. checked

497 Sims D. Marketing Report Uzbekistan. Field Visit 6/10/93 to 15/10/93. [15/10/93]. Guildford Depository. BAT. Bates No: 203465873-93.

377

498 Whitehouse A. Belarus. 14/06/1995. Guildford Depository. BAT. Bates No: 203476989-95.

499 British American Tobacco. Joint venture with tobacco Factory Grodno. 08/03/1995. Guildford Depository. BAT. Bates No: 203476804.

500 British American Tobacco. Summary of BAT's proposals for the modernisation of Tobacco factory Grodno and benefits to Belarus. 06/03/1995. Guildford Depository. BAT. Bates No: 203476805.

501 Whitehouse A. Note of a lunch held at BAT with MR Yaroshuk of the Belarusian Ministry of Agriculture on 27th January 1995. 27/01/1995. Guildford Depository. BAT. Bates No: 203477075-6.

502 Hewko J. Memorandum. 20/02/1995. Guildford Depository. BAT. Bates No: 203476839-40.

503 Oltermann G. Report on the visit to Grodno on 28/29th March 1995. 31/03/1995. Guildford Depository. BAT. Bates No: 203477015-20.

504 Coe L. Note to Anne Johnson Belarus. 20/04/1995. Guildford Depository. BAT. Bates No: 203477004.

505 Lioutyi A. Contact with Minister Leonov. 07/07/1995. Guildford Depository. BAT. Bates No: 203476987.

506 British American Tobacco. [Part of review of Moldovan Tobacco industry]. Guildford Depository. BAT. Bates No: 602750504-14.

507 Taylor R. SUBJECT: visit of Moldovan Minister of Agriculture to Brazil. 21/05/1993. Guildford Depository. BAT. Bates No: 602751754.

508 Bruce-Miller N. Uzbekistan - File note - Chairman's visit 14-16/12/93. 17/12/1993. Guildford Depository. BAT. Bates No: 203835033-4.

509 BAT. Press Announcement. BAT Investment in Uzbekistan. 16/05/1994. Guildford Depository. BAT. Bates No: 503801058-9.

378

510 Sheehy P. [Letter to President Kravchuk]. 07/10/1992. Guildford Depository. BAT. Bates No: 203837060-1.

511 European Bank of Reconstruction and Development, trade exclusion list. http://www.ebrd.com/about/index.htm (last accessed 6 May 2003)

512 Salter R. EBRD. 24/10/91. BAT. Access Date: 30/06/2003. Bates No: 201829883. URL: http://www.library.ucsf.edu/tobacco/batco/html/14200/14261/otherpages/2.html

513 British American Tobacco. Uzbekistan - Cigarette Market. 12/01/1994. Guildford Depository. BAT. Bates No: 203835011-8.

514 Brookes N. Excise issues. 02/12/1992. Guildford Depository. BAT. Bates No: 203475147.

515 British American Tobacco. (Secret) Ukraine Business Plan for joint enterprise between BAT Industries Plc and the Cherkassy, Prilucky and Monastirisk Factories. 24/05/1992. Guildford Depository. BAT. Bates No: 301735233-72.

516 Bruce-Miller N. [letter re visit to Pecs factory]. 10/02/1994. Guildford Depository. BAT. Bates No: 301733123.

517 Herter U. Keynote speech to the World Tobacco Symposium Wednesday 22nd September 1993. Ulrich Herter, Managing Director Tobacco BAT Industries plc. 22/09/1993. Guildford Depository. BAT. Bates No: 203831337-47.

518 British American Tobacco. BAT Industries PLC Reporting Minutes. Secret. 19/09/1994. Guildford Depository. BAT. Bates No: 203470846-47.

519 Brookes N. Moldova. 07/06/1994. Guildford Depository. BAT. Bates No: 602750624.

520 Taylor R. Moldova - Leaf purchases by BAT. 07/06/1994. Guildford Depository. BAT. Bates No: 602750622.

521 Brady B. Visit Notes from St Petersburg, Hamburg and Beld (Slovenia). 07/02/1994. Guildford Depository. BAT. Bates No: 502649219-20.

379

522 British American Tobacco. Letter of Intent between Ukrtabakprom and the Prilucky, Cherkassy and Monastirisk factories (hereafter together called "the Ukrainian Partners") and BAT Industries plc (hereinafter called “BAT”). 24/02/1992. Guildford Depository. BAT. Bates No: 203836896-900.

523 Johnson A. Meetings in Kiev 23rd & 24th September 1992. 29/09/1992. Guildford Depository. BAT. Bates No: 203837260-5.

524 British American Tobacco. Memorandum of Understanding between The State Committee of State Property and Privatisation of the Republic of Uzbekistan, the Uzbek State Joint Stock Concern for the Food Industry of the Republic of Uzbekistan (together, the "Uzbekistan Parties") and BAT Industries plc. 29/09/1993. Guildford Depository. BAT. Bates No: 203835130-7.

525 British American Tobacco. Note to the CEC. Uzbekistan. 09/06/1993. Guildford Depository. BAT. Bates No: 203832650-7.

526 Sheehy P. [Letter to President Nazerbayev]. 18/01/1993. Guildford Depository. BAT. Bates No: 203461299-300.

527 Bruce-Miller N. Kazakhstan. 05/02/1993. Guildford Depository. BAT. Bates No: 203461266-9.

528 Van Waay A, Booker RMK. Secret. Report Alma Ata, Kazakhstan 23rd-27th November 1992.. 02/12/1992. Guildford Depository. BAT. Bates No: 600001059-67.

529 Brookes N. [Fax from N Brookes re Kazakhstan]. 05/02/1993. Guildford Depository. BAT. Bates No: 203461270.

530 TAS (TAS) - Marlboro cigarette production stops in Samara. 21/01/1993. Guildford Depository. BAT. Bates No: 203461271.

531 Unknown . Anon. Taking Stock. Geoffrey C. Bible outlines Philip Morris strategy - past, present and future - for New York analysts. Tobacco Reporter. [01/04/1994]. Guildford Depository. BAT. Bates No: 203459210-2.

532 Taylor R. [Opportunities in CEE and other newly opened markets]. 25/03/1993. Guildford Depository. BAT. Bates No: 602750630.

380

533 Taylor R. [Opportunities in CEE and other newly opened markets]. 25/03/1993. Guildford Depository. BAT. Bates No: 602750631.

534 Taylor R. [Opportunities in CEE and other newly opened markets]. 25/03/1993. Guildford Depository. BAT. Bates No: 602750632.

535 Brookes N. New Business Development Monthly Report to Mid-December 1992. 24/12/1992. Guildford Depository. BAT. Bates No: 203481267-73.

536 Van Waay A. Kazakhstan. 16/12/1992. Guildford Depository. BAT. Bates No: 203461354.

537 Van Waay A. Executive Summary. Visit to Alma Ata, Kazakhstan. 03/12/1992. Guildford Depository. BAT. Bates No: 203461361-6.

538 British American Tobacco. Proposal for the basis for future co-operation between BAT and AO Donskoy Tabak. 05/07/1993. Guildford Depository. BAT. Bates No: 203466258-63.

539 Wells W. Note of a lunch with Dr Khamidov of UPP at Windsor House of Tuesday 23rd November 1993 at 1245 hrs. 23/11/1993. Guildford Depository. BAT. Bates No: 301728270-3.

540 Watterton D. Secret. New Business Development monthly report - August 1992. 04/09/1992. Guildford Depository. BAT. Bates No: 203471370-5.

541 Various. [Details of Philip Morris’s proposal for Kazakhstan]. 25/03/1993. Guildford Depository. BAT. Bates No: 203461429-70.

542 Christian Aid. Hooked on Tobacco. Report by Christian Aid/ DESER on British American Tobacco subsidiary, Souza Cruz. 2002. (http://www.christian- aid.org.uk/indepth/0201bat/ last accessed 3/3/03)

543 Malone RE. Tobacco industry surveillance of public health groups: the case of STAT and INFACT. American Journal of Public Health; 2002; 92:955-60.

544 Philip Morris. List of tobacco control activists. Bates No.: 2081380266-73 URL: http://legacy.library.ucsf.edu/cgi/getdoc?tid=ykw65c00&fmt=tif&ref=results

381

545 Wood M. Eastern Europe: A new market ripe for exploitation. [1991]. Guildford Depository. BAT. Bates No: 203825529-30.

546 Whitehouse A. Note of meetings held with Salans Herzfeld & Heilbronn in Warsaw on December 3rd and 4th 1992. 06/12/1992. Guildford Depository. BAT. Bates No: 203842115-8.

547 Burrill DM. Ukraine - Procurator's view. 06/07/1993. Guildford Depository. BAT. Bates No: 301772763-6.

548 British American Tobacco. Living and Working in Moscow City. [01/09/1992]. Guildford Depository. BAT. Bates No: 203481072-106.

549 Organisation for Economic Cooperation and Development. OECD Guidelines for Multinational Enterprises: Text, Commentary and Clarifications (October 2001) DAFFE/IME/WPG(2000)15/FINAL, 2001. http://www.olis.oecd.org/olis/2000doc.nsf/4f7adc214b91a685c12569fa005d0ee7/d1bada 1e70ca5d90c1256af6005ddad5/$FILE/JT00115758.PDF (last accessed 6 April 2005)

550 Warner KE. The economics of tobacco: myths and realities. Tobacco Control 2000;9: 78-89.

551 Lightwood J, Collins D, Lapsley H, Novotny TE. Estimating the costs of tobacco use. In: Tobacco Control in Developing Countries. Jha P, Chaplouka F (eds.). Oxford: Oxford University Press, 2000.

552 Jacobs R, Gale HF, Capehart TC, Zhang P, Jha P. The supply-side effects of tobacco control policies. In: Tobacco Control in Developing Countries. Jha P, Chaplouka F (eds.). Oxford: Oxford University Press, 2000.

553 Golden leaf, barren harvest: the costs of tobacco farming, Washington: Campaign for Tobacco Free Kids, 2001.

554 Ministry of External Trade and Industry, Kyrgyz Republic. Trade balance for 2000. http://www.mvtp.kg/english/analysis/report2000_1.html (last accessed 21 May 03)

555 United Nations Food and Agriculture Organisation. FAOSTAT agricultural data. http://apps.fao.org/page/collections?subset=agriculture (last accessed 21 May 2003)

382

556 Krasovsky K, Andreeva T, Krisanov D, Mashliakivsky M, Rud G (for the Alcohol and Drug Information Center Ukraine). Economics of tobacco control in Ukraine from the public health perspective. Kiev: Polygraph Center TAT, 2002.

557 [Espindola E.] [Details of meetings in Kazakhstan]. [26/11/1992]. Guildford Depository. BAT. Bates No: 203461328-34.

558 Yurekli A. Uzbekistan Health II Project. Report on tobacco control in Uzbekistan April 6-16, 2005.

559 Ellison M. Tobacco giant apologises to Czechs. The Guardian, 27th July 2001. (http://society.guardian.co.uk/cancer/story/0,8150,528357,00.html last accessed 3/3/03)

560 Bates C. Study shows that smoking costs 13 times more than it saves Br Med J 2001;323:1003.

561 Collin J, Gilmore A. Corporate (Anti)social (ir)responsibility: transnational tobacco companies and the attempted subversion of global health policy. Global Social Policy 2002; 2(3): 354-60.

562 European Parliament. Directorate-General for Research. Note on the political and economic situation in Moldova and its relations with the European Union. (IV/WIP/2002/07/0020-0021). Luxembourg: European Parliament, 16th July 2002.

563 Radziwill A, Petrishin O. GDN Country Growth Study. Barriers to growth in Moldova. Economics Education and Research Consortium, 2001. www.eerc.ru/activ/projects/GRP/papers/04_Moldova.pdf

564 British American Tobacco. BAT Industries PLC Reporting Minutes. Secret. 05/12/1994. Guildford Depository. BAT. Bates No: 203470836-7.

565 Economist Intelligence Unit. Reemtsma's $59 million bid wins control of Moldova tobacco industry. 21/9/1996. Bates Number: 2062387407A http://legacy.library.ucsf.edu/cgi/getdoc?tid=jzj93c00&fmt=pdf&ref=results.

566 Reuters. Germany's Reemtsma dismayed by Moldova tobacco move. 02/12/1997. Bates Number: 2062386101A. URL:

383

http://legacy.library.ucsf.edu/cgi/getdoc?tid=kle48d00&fmt=pdf&ref=results (last accessed 19 July 2004)

567 Tobacco Documentation Centre. Smoking issues status book bulletin on proposed, pending and recently approved legislation. 1997. Bates Number: 2064324602/4604. URL: http://legacy.library.ucsf.edu/tid/wmt08d00 (last accessed 19 July 2004)

568 Letter of Intent and Memorandum of Economic and Financial Policies of the government of Moldova, July 1999. http://www.imf.org/external/np/loi/1999/072999.htm (last accessed 8 July 2003)

569 Dow Jones Newswires. IMF: Moldova must privatize wine, tobacco. November 3, 1999.

570 Dow Jones Newswires. IMF Threatens To Cut Relations With Moldova. March 24, 2000

571 Dow Jones Newswires. IMF freezes financial support planned for Moldova. November 5 1999.

572 The Economist. Can Moldova get worse? 13th June 2000.

573 Department of Privatization of the Republic of Moldova. http://www.privatization.md/privatization/privatization_companies/strategic_projects/wi ne_tobacco_energy/ (last accessed 7 July 2003)

574 International Monetary Fund. Moldova - Interim Poverty Reduction Strategy Paper, April 21 2001. http://www.imf.org/external/np/prsp/2002/mda/01/042102.pdf (last accessed 27 August 2003)

575 Communists devastate the Tobacco Enterprise in Chisinau. Flux. November 21 2000. (IMF translation)

576 Costas N. Mafia tutunului. Jurnal de Chisinau, 2003, 28 February, p.10.

577 Costas N. Mafia tutunului-2. - Jurnal de Chisinau, 2003, 21 March, p.3.

578 Poiana M. 80.000 USD pentru plaisirul lui Misin-junior. Accente 30 May 2002.

384

http://www.accente.com.md/nr46/p1.shtml (last accessed 8 July 2003)

579 Freedom House. Nations in Transition Country Report: Moldova. http://www.freedomhouse.org/research/nattransit.htm (last accessed 8 July 2003)

580 Murphy M. Moldova - Initial investment appraisal for leaf processing. 30/09/1993. Guildford Depository. BAT. Bates No: 602750677-81.

581 Taylor R. Moldova : Domestic tobacco usage and leaf exports. 04/03/1994. Guildford Depository. BAT. Bates No: 602750570-92.

582 Lioutyi A. Trip to Moldova 28-31 May. 3/6/96. Minnesota Depository. BAT. Bates no. 700614199-201.

583 Cozhokaru D. BAT team visit to Kishinev. 22/10/93. Guildford Depository. BAT. Bates No: 301771789.

584 Kirke J. Moldova visit report 23-27 October. 27/10/93. Guildford Depository. BAT. Bates No: 301771798-801.

585 Taylor R. 1993 Crop Financing Proposal. Meeting with KTK 04/06/93. 6/5/93. Guildford Depository. BAT. Bates No: 301771842-4.

586 BAT. Visit Notes Moldova - 14-16th June 1993. 14/6/93. Guildford Depository. BAT. Bates No: 301771753-65.

587 Taylor R. [Letter re machinery purchased by BAT]. 25/10/93. Guildford Depository. BAT. Bates No: 301771775-6.

588 Kirke J. Confidential for BAT eyes only. Current reaction to "Moldova: Tobacco Industry Development Plan." as at the 27th October. 27/10/93. Guildford Depository. BAT. Bates No: 301771792-5.

589 Webb,W. Kazakhstan - cigarette manufacturing investment in Almaty - $360 million. 16/6/1996. Bates Number: 2048037644/7660. URL: http://legacy.library.ucsf.edu/cgi/getdoc?tid=ecq92e00&fmt=pdf&ref=results (last accessed 19 July 2004)

385

590 Melling PJ. Moldova. 21/7/95. Minnesota Depository. BAT. Bates no. 800198643-5.

591 Stakes have been made, gentlemen! (Ставки сделаны, господа) Interview by Yacov Gurevici with Mr. Yacov Mogoreanu. Weekly “Kishinevskie novosti” (“Chisinau news”), June 15, 1996.

592 Lioutyi A. Trip to Ukraine 1-3 July 1996. July 1996. Minnesota Depository. BAT. Bates no: 700587571-4.

593 Jamieson B. Moldova, republic of corruption. Sunday Telegraph 29th September 1996. http://www.telegraph.co.uk/htmlContent.jhtml?html=%2Farchive%2F1996%2F09%2F2 9%2Fwmold29.html (last accessed 17 July 2003)

594 Taylor R. Moldova - Revised tobacco production and leaf processing programme. 24/05/1994. Guildford Depository. BAT. Bates No: 602750636-46.

595 Zatusevski I. Confronting tobacco companies promotion actions. European Conference on Tobacco or Health. Warsaw, 2002.

596 Flux. KP v Moldove. 25 April, p.3 Logos-press, 2003. (see also: 25 April, p.21; 23 May, p.3; 8 June, p.30;20 June, p.3 etc).

597 Royale and Brooklyn made in Moldova. Interview with general director of Tutun-CTC, Petru Poian. Observator, April 2001. http://www.observator.mdl.net/reviste/nr29/eng_tobacco.htm (last accessed 7 July 2003)

598 Imperial Tobacco Group. Annual report and accounts 2004. http://www.imperial- tobacco.com/files/financial/reports/ar200/files/annual_report_2004.pdf (accessed 29 March 2005)

599 Eaglesham J, Blackwell D. Imperial Tobacco cuts cigarette exports. Financial Times 7 March 2003.

600 British Helsinki Human Rights Group (2002) BAT in Uzbekistan 2002. http://www.bhhrg.org/CountryReport.asp?CountryID=23&ReportID=5&ChapterID=31 &next=next&keyword= (accessed 25 July 2005)

386

601 Ministry of Public Health, Thailand. Thailand Health Profile 1999-2001.Chapter 4. http://www.moph.go.th/ops/thealth_44/CHA4_7.PDF (last accessed 6th April 2004)

602 Price Waterhouse Coopers. Uzbekistan a business and investment guide. March 2004.

603 Monbiot G. Britain and the US claim a moral mandate – and back a dictator who boils victims to death. The Guardian Tuesday October 28 2003.

604 Cohen N. Trouble in Tashkent. The West – and particularly the US – seems blind to Uzbekistan’s cruel government. The Observer 15th December 2002.

605 Paton Walsh W. Abuses force America to end aid to Uzbekistan. The Guardian Saturday January 17 2004.

606 Wagstyl S and Gorst I. UN demands independent probe into Uzbek killings. Financial Times. Wednesday 18 May 2005.

607 Select Committee on Foreign Affairs Minutes of Evidence. Memorandum submitted by the Foreign and Commonwealth Office. March 1999. http://www.parliament.the-stationery- office.co.uk/pa/cm199899/cmselect/cmfaff/349/9052502.htm (last accessed 15 November 2004)

608 Herter U. Note to the Chief executives committee. Uzbekistan. 14/10/1994. Guildford Depository. BAT. Bates No: 503800815-29.

609 Ellman M. The Increase in Death and Disease under "Katastroika.". Cambridge J Econ 1994; 18: 329-55.

610 The World Trade Organization. Trade and competition policy. http://www.wto.org/english/thewto_e/minist_e/min99_e/english/about_e/16comp_e.htm (accessed 18 July 2005)

611 Pass C, Lowes B, Pendleton A, Chadwick L, O’Reilly D, Afferson M. Dictionary of Business. Third edition. Glasgow: Collins, 2002.

612 Institute Of International Business Law And Practice. Due process and anti- competitive practices. London: International Chamber of Commerce, 1994.

387

613 Broadman HG. Policy Research Working Paper 2331. Competition, Corporate Governance, and Regulation in Central Asia. Uzbekistan’s Structural Reform Challenges. Washington DC: World Bank, May 2000.

614 British American Tobacco. Standards of Business Conduct, 2003. http://www.bat.com/oneweb/sites/uk__3mnfen.nsf/0/655fc96e7131799880256df300532 020?OpenDocument (last accessed 9th July 2004)

615 Smith E A, Malone R E.Thinking the "unthinkable": why Philip Morris considered quitting. Tob. Control 2003; 12: 208 - 213.

616 Unknown . Draft 2, Project Umpire, Skeleton business Plan. [01/02/1994]. Guildford Depository. BAT. Bates No: 600507395-420.

617 Wells W. Preliminary briefing note. 03/08/1993. Guildford Depository. BAT. Bates No: 203832525-8.

618 van Waay A. Note of World Economic Forum special trip to Central Asia. 27/04/1993. Guildford Depository. BAT. Bates No: 203459475-81.

619 van Waay A. [Letter to Mr Ysupov, Ministry for Foreign Economic Relations]. 28/04/1993. Guildford Depository. BAT. Bates No: 203459501-2.

620 Herter U, Brookes N. Extract of Minutes of meeting of the Chief Executive's Committee held on Thursday, 17th June 1993. 17/06/1993. Guildford Depository. BAT. Bates No: 503801138-9.

621 Herter U. Note to the CEC: Uzbekistan. Secret. 10/06/1993. Guildford Depository. BAT. Bates No: 203832336-40.

622 Herter U. Note to the CEC: Uzbekistan. Secret. 05/07/1993. Guildford Depository. BAT. Bates No: 203832600-2.

623 Herter U. Note to the CEC: Uzbekistan. Secret. 05/07/1993. Guildford Depository. BAT. Bates No: 203835128-9.

624 Herter U. Note to the Chief Executives Committee. Central Asian Republics. 10/09/1993. Guildford Depository. BAT. Bates No: 503801105-9.

388

625 Unknown . BATs strategy for Central Asia. 03/09/1993. Guildford Depository. BAT. Bates No: 503801116-9.

626 Rowlands-Rees G. Uzbekistan visit 26-30 July 1993. A discussion note.. 02/08/1993. Guildford Depository. BAT. Bates No: 203831285-93.

627 Wells W. Project Umpire. Note of meeting with Guinora Mahmudova of the GKI at 1600hrs on Monday 23rd August, 1993 in Tashkent.. 25/08/1993. Guildford Depository. BAT. Bates No: 203832500-2.

628 Kirillova E. BAT-Uzbekistan. 20/07/1993. Guildford Depository. BAT. Bates No: 203832597-9.

629 Wells W. [Letter to Ms Mahmudova]. 12/08/1993. Guildford Depository. BAT. Bates No: 203832595-6.

630 Wells W. Project Umpire. 13/08/1993. Guildford Depository. BAT. Bates No: 203832577-8.

631 Bruce-Miller N. [Draft letter to Dr Gafurov, GKI, Uzbekistan]. 13/08/1993. Guildford Depository. BAT. Bates No: 203832579-4.

632 Wells W. [admin letter]. 20/08/1993. Guildford Depository. BAT. Bates No: 203832503-5.

633 BAT Industries. Memorandum of Understanding between the Uzbek State Joint Stock Concern for the Food Industry of the Republic of Uzbekistan, the State Committee of State Property and Privatisation of the Republic of Uzbekistan and BAT Industries plc of London, England.. 5/10/93. Guildford Depository. BAT. Bates No: 301731698-706.

634 British American Tobacco. Note of a lunch with Dr Khamidov of UPP at Windsor House of Tuesday 23rd November 1993 at 1245 hrs. 23/11/1993. Guildford Depository. BAT. Bates No: 301728270-3.

635 Wells W. Project Umpire. Note of meeting help at UPP on Monday 15th November, 1993 at 1615 hrs.. 29/11/1993. Guildford Depository. BAT. Bates No: 301728265-7.

389

636 Wells W. Negotiations with BAT Industries. 21/09/1993. Guildford Depository. BAT. Bates No: 203832436-9.

637 Wells W. Project Umpire: Notes from visit to Tashkent, Tuesday 11th to Friday 14th January 1994. 20/01/1994. Guildford Depository. BAT. Bates No: 600507380-5.

638 BAT Uzbekistan (Investments) Ltd. Registered No. 2906405. Directors’ report and accounts for the years ended 31 December 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002 and 2003. (All nine reports obtained from Companies House http://direct.companieshouse.gov.uk/ 8 June 2005)

639 Herter U, Brookes N. Extracts of Minutes of meeting of the Chief Executives Committee held on Monday 13th Sepember 1993. 13/09/1993. Guildford Depository. BAT. Bates No: 503801104.

640 J Henry Schroder Wagg & Co. Ltd. BAT Industries. Project Umpire. Administrative Documentation. 04/11/1993. Guildford Depository. BAT. Bates No: 301728279-95.

641 Wells W. Project Umpire: issues for Dr. Khamidov et al and Revised draft of BAT's proposed investment in Uzbekistan. Issues for discussion. 29/11/1993. Guildford Depository. BAT. Bates No: 301728237-41.

642 House of Commons Hansard Debates for 10 Dec 1993 (http://www.parliament.the-stationery-office.co.uk/pa/cm199394/cmhansrd/1993-12- 10/Writtens-3.html) last accessed 15th April 2004.

643 Select Committee on Foreign Affairs. Minutes of Evidence. ANNEX A8 Uzbekistan. http://www.publications.parliament.uk/pa/cm199899/cmselect/cmfaff/349/9052510.htm (last accessed 1 April 2005)

644 BAT. Note of meeting held at 1430 hrs on Monday, 20th December, 1993 at Windsor House. 20/12/1993. Guildford Depository. BAT. Bates No: 203835026-32.

645 Wells W. Protocol of Intent: the Republic of Uzbekistan and BAT Industrues plc. 16/12/1993. Guildford Depository. BAT. Bates No: 203835020-5.

390

646 Herter U. Note to the Chief Executives Committee. Uzbekistan. 28/01/1994. Guildford Depository. BAT. Bates No: 503801081-3.

647 Karimov I. Decision of the Cabinet of Ministers of the Republic of Uzbekistan. 11/02/1994. Guildford Depository. BAT. Bates No: 503801074.

648 British American Tobacco. Investment proposal by the Uzbek State Joint Stock Concern for the food industry of the Republic of Uzbekistan ("Uzpischeprom") and the BAT Group of Companies ("BAT") for the establishment and development of a joint venture. 25/03/1994. Guildford Depository. BAT. Bates No: 203459327-43.

649 Saldutti J. [Letter re UZBAT joint venture]. 12/04/1994. Guildford Depository. BAT. Bates No: 600507809-15.

650 Wells W. BAT Industries plc ("BAT")'s proposed Joint Venture investment ("UZBAT") in the Republic of Uzbekistan. 14/04/1994. Guildford Depository. BAT. Bates No: 600507795-808.

651 BAT. BAT: Project Umpire. 25/10/1993. Guildford Depository. BAT. Bates No: 301728296-307.

652 BAT. Preliminary draft: 23rd november 1993 "Strategic Vision" agreement. 23/11/1993. Guildford Depository. BAT. Bates No: 301728274-6.

653 Bruce-Miller N. [hand written note on Deam Sim's visit]. 14/10/1993. Guildford Depository. BAT. Bates No: 203831279-80.

654 Bruce-Miller N. Re: Marketing strategy - Central Asia. 03/12/1993. Guildford Depository. BAT. Bates No: 203459271-3.

655 Bruce-Miller N. Note to Mr UGV Herter. Uzbekistan. 28/10/1993. Guildford Depository. BAT. Bates No: 203459276.

656 Wells W. Project Umpire: Fiscal and Excise Working Groups. 19/04/1994. Guildford Depository. BAT. Bates No: 600507774-5.

657 Unknown . Project Umpire, Fiscal Working Group Meeting, Thurdsay April 21st at 3pm. 21/04/1994. Guildford Depository. BAT. Bates No: 600507769-70.

391

658 King J. Project Umpire. 21/06/1994. Guildford Depository. BAT. Bates No: 600507652-5.

659 Norwell I. BAT investment in Uzbekistan. 16/05/1994. Guildford Depository. BAT. Bates No: 500852311-20.

660 Bishop D. Uzbekistan - Visit 20/21 June. 22/06/1994. Guildford Depository. BAT. Bates No: 600507656-8.

661 Marr M, Sims D. Uzbekistan Marketing Plan 1994-2003. Discussion draft 3. [01/11/1993]. Guildford Depository. BAT. Bates No: 203459117-35.

662 Herter U. The finalisation of BAT's planned investment in Uzbekistan. 08/07/1995. Guildford Depository. BAT. Bates No: 600507665-7.

663 Buckley N. BAT-man's opening. 19/07/1994. Guildford Depository. BAT. Bates No: 600507621.

664 Wells W. Project Umpire. Uzbek Health Ministry Decree promoting the restriction of smoking. 25/08/1994. Guildford Depository. BAT. Bates No: 500852348- 9.

665 Wilson M. BAT Industries PLC Reporting Minutes. Secret. 05/09/1994. Guildford Depository. BAT. Bates No: 203470848-9.

666 Wells W. Project Umpire, Import Duties. 24/06/1994. Guildford Depository. BAT. Bates No: 600507677-9.

667 Wells W. Project Umpire. 07/10/1994. Guildford Depository. BAT. Bates No: 503800851-2.

668 Wells W. URGENT PROJECT UMPIRE COMPLETION. 26/10/1994. Guildford Depository. BAT. Bates No: 600507538-45.

669 Anon. BAT Bulletin. News: BAT takes the lead in Uzbekistan. August 1995, Bates No. 500161238.

670 Anon. BAT Bulletin. High profile visit to Uzbekistan. February 1995. Guildford Depository. Bates no.: 500161154.

392

671 Bryden I. [letter from Chase Manhattan Bank re payment for Uzbek operations]. 23/11/1994. Guildford Depository. BAT. Bates No: 503800810.

672 BAT Uzbekistan (Investments) Limited. Report and accounts for the year ended 31 December 1994. Registered no. 2906405. (Obtained from Companies House http://direct.companieshouse.gov.uk/ on 9/6/06)

673 Unknown . [Sir Patrick Sheehy in Uzbekistan. Article in Business briefing BAT corporate newsletter]. [01/01/1995]. Guildford Depository. BAT. Bates No: 500057716.

674 British American Tobacco. Investing in Uzbekistan. http://www.bat.com/oneweb/sites/uk__3mnfen.nsf/0/197ac91697945f5080256bf400019 8f9?OpenDocument (last accessed 20/7/04)

675 Anon. BAT Output. Tobacco Journal International 2000; March. http://www.tobaccojournal.com/show_artikel.php3?id=1565&banner=etm1 (last accessed 14 April 2005)

676 BAT Uzbekistan cigarette output down 7.5% in 2003 http://www.bisnis.doc.gov/bisnis/bisdoc/0402UZ_CommercialReport.htm (last accessed 5 April 2005)

677 American Stock Exchange. British-Uzbek firm cuts cigarette output by over 4 per cent.. 16 November 2004. http://www.amex.com/?href=/newsDetails/CmnNewsDet.jsp?id=XpressFeed_NewsDeta ils_1100616926540.html (last accessed 3 April 2005)

678 BAT Uzbekistan predicts 2004 net losses of $12.5 million. http://www.tobacco.org/news/175308.html (last accessed 3 April 2005)

679 BAT Uzbekistan (Investments) Ltd. Registered No. 2906405. Directors’ report and accounts for the year ended 31 December 2002. (Obtained from Companies House http://direct.companieshouse.gov.uk/ 8 June 2005)

680 BAT Uzbekistan (Investments) Ltd. Registered No. 2906405. Directors’ report and accounts for the year ended 31 December 2003. (Obtained from Companies House http://direct.companieshouse.gov.uk/ 8 June 2005)

393

681 The price is not quite right. The Economist, 5 July 2001 http://www.ash.org.uk/html/conduct/html/pricefixing.html [Accessed 19 June 2005].

682 Republic of Uzbekistan. Law on Foreign Investment (Amended 2.07.92). BAT. Bates No: 301781981-1992 http://bat.library.ucsf.edu/tid/ufu70a99

683 Yurekli AA. De Beyer J. Did the entry of private cigarette producers in Turkey and Ukraine increase cigarette consumption? Unpublished report.

684 Hellman JS. Jones G, Kaufmann D. Far from Home. Do foreign investors import higher standards of governance in transition economies. Draft for discussion. Washington DC: World Bank, August 2002.

685 British American Tobacco Uzbekistan. Social Report 2002/3. Tashkent: BAT Uzbekistan, 2003. http://www.bat.com/oneweb/sites/uk__3mnfen.nsf/0/4F88F56106625FE5C1256E99002 96A91?opendocument (last accessed 24 July 2005)

686 http://www.weforum.org/ (last accessed 6 April 2005)

687 Smith R. Economics first; health third, fourth, or nowhere. BMJ 2004; 328: 0.

688 Select Committee on Foreign Affairs. Appendices to the Minutes of Evidence – Sixth Report. Appendix 13. Memorandum submitted by British American Tobacco. http://www.parliament.the-stationery- office.co.uk/pa/cm199899/cmselect/cmfaff/349/349ap20.htm (last accessed 6 April 2005)

689 WHO Tobacco Free Initiative. Building blocks for tobacco control: A handbook. Geneva: World Health Organization. 2004.

690 Townsend J. The Role of Taxation in Tobacco Control. In: Abedian I, van der Merwe R, Wilkins N, Jha P. eds. The Economics of Tobacco Control: Towards an optimal policy mix (Cape Town, SA: Applied Fiscal Research Centre, University of Cape Town). 1998

691 Warner K. The Economics of Tobacco and Health: an Overview. n: Abedian I, van der Merwe R, Wilkins N, Jha P. eds. The Economics of Tobacco Control: Towards

394

an optimal policy mix (Cape Town, SA: Applied Fiscal Research Centre, University of Cape Town). 1998

692 Ranson K, Jha P, Chaloupka F and Nguyen S. The effectiveness and cost- effectiveness of price increases and other tobacco control policies. In Jha P and Chaloupka F eds. Tobacco Control in Developing Countries (Oxford: Oxford University Press)

693 World Health Organization. The World Health Report 2003: Shaping the future (Geneva: WHO)

694 BATCo 1993-1997 Company Plan. Feb 1993. Guildford Depository. Bates No: 500008637-8718. British American Tobacco Documents Archive. http://bat.library.ucsf.edu/data/a/c/m/acm70a99/acm70a99.pdf (accessed 11 April 2005)

695 Bishop D. Key area paper: indirect taxation. BAT. March 1994. Bates No.: 503911771-503911780. British American Tobacco Documents Archive. http://bat.library.ucsf.edu/data/h/f/k/hfk11a99/hfk11a99.pdf (accessed 11 April 2005)

696 Selby J, Bishop D. Proposal on Excise taxation for Imported Cigarettes into Uzbekistan. 15/02/1995. Guildford Depository. BAT. Bates No: 600507451-9.

697 Reuters. Uzbekistan: Uzbekistan and Kazakhstan to form Economic Union. [10/01/1994]. Guildford Depository. BAT. Bates No: 503801161-2.

698 Wells W. Projects Umpire and Kant, Final notes of visit to Central Asia, 15th – 19th November, 1993. 17/12/1993. Guildford Depository. BAT. Bates No: 203835042- 61.

699 Dufty C. Uzbekistan - cigarette indirect tax treatment. 22/04/1994. Guildford Depository. BAT. Bates No: 600507764-6.

700 Bishop D. Uzbekistan - Tobacco taxes, visit by DJ Bishop and CD Dufty - 10-14 January 1994. 19/01/1994. Guildford Depository. BAT. Bates No: 600507821-7.

701 Wells W. Uzbekistan: Draft letter to the Dep. Minister of Tax Inspection. 23/12/1993. Guildford Depository. BAT. Bates No: 203834999-5002.

395

702 Bishop D. Uzbekistan - Excise. 31/01/1994. Guildford Depository. BAT. Bates No: 600507829.

703 Dufty C. Uzbekistan - indirect tax meeting with Ministry of Finance. 03/03/1994. Guildford Depository. BAT. Bates No: 600507694-7.

704 Wells W. Project Umpire: Excise and VAT reform proposals. 09/06/1994. Guildford Depository. BAT. Bates No: 600507755.

705 Bishop D. Reference our conversation today re level playing field or protective tariffs. 22/06/1994. Guildford Depository. BAT. Bates No: 600507699.

706 British American Tobacco. Cigarette Prices in Uzbekistan and adjoining countries. 20/06/1994. Guildford Depository. BAT. Bates No: 600507709-11.

707 Bishop D. Meeting at Tashkent Ministry of Finance, 13 July 1994. 19/07/1994. Guildford Depository. BAT. Bates No: 503922515-7.

708 Bishop D. Re: UZBAT AO. 01/07/1994. Guildford Depository. BAT. Bates No: 503922519-20.

709 Bishop D. Uzbekistan. 20/07/1994. Guildford Depository. BAT. Bates No: 503922514.

710 Schroders. Project Umpire. Meeting to be held at Windsor House on 23rd August at 1430hrs Proposed agenda. 23/08/1994. Guildford Depository. BAT. Bates No: 503800858-9.

711 The World Tobacco File 1998 Volume One. London: Market Tracking International Ltd.

712 Bishop D. Meeting - Ministry of Finance. 19/07/1994. Guildford Depository. BAT. Bates No: 503922518.

713 Wells W. Project Umpire: Current o/s position. 20/09/1994. Guildford Depository. BAT. Bates No: 600507555-61.

714 Selby J. Subject: Excise Proposal - urgent. [15/02/1995]. Guildford Depository. BAT. Bates No: 600507450.

396

715 Selby J. Excise Proposals. 09/02/1995. Guildford Depository. BAT. Bates No: 600507500-1.

716 Selby J. Subject: excise - Urgent. 31/01/1995. Guildford Depository. BAT. Bates No: 600507514.

717 Lee K, Gilmore A, Collin J. Breaking and entering: BAT’s strategy to re-enter China from the late 1970s to mid 1990s. Tobacco Control 2004; 13 Supp II: ii88-95.

718 Collin J, LeGresley E, MacKenzie R, Lawrence S, Lee K Complicity in Contraband: British American Tobacco and Cigarette Smuggling in Asia,” Tobacco Control, 2004,13(Supp II): ii96-ii111.

719 Herter U. Note to the TSG 1c./11.5.94. Central Asia and the Caucasus. 4/5/94. Guildford Depository. BAT. Bates No: 500036164-6.

720 Uzbekistan Information Directory. Tobacco consumption in Uzbekistan increasing over 7 per cent a year, May 31 2002. http://uzland.narod.ru/2002/june/03/05.htm (last accessed July 04)

721 British American Tobacco. New Tobacco markets. 3/6/92. Guildford Depository. BAT. Bates No: 203469395-403.

722 Van Waay A. Ukraine: Marketing section. 05/03/1992. Guildford Depository. BAT. Bates No. 301689862-4

723 Unknown. (Secret) Ukraine Business Plan for Joint enterprise between BAT Industries Plc and the Cherkassy, Prilucky and Monastirisk Factories. 24/05/1992. Guildford Depository. BAT. Bates No: 301735258-72.

724 British American Tobacco. BAT Industries PLC Reporting Minutes. Secret. 19/09/1994. Guildford Depository. BAT. Bates No: 203470846-47.

725 British American Tobacco. The Appendix. Investment Assumptions and Conditions. 11/11/1994. Guildford Depository. BAT. Bates No: 600500268.

726 Bishop D. Kyrgyzstan - notes for presentation to Ministry of Finance. 18/11/1994. Guildford Depository. BAT. Bates No: 600500226-9.

397

727 Bishop D. Re: Kyrgyzstan. 21/11/1994. Guildford Depository. BAT. Bates No: 600500220.

728 BAT Industries plc. Investment proposal by BAT Industries plc for an investment in Bishkek tobacco factory and Bishkek fermentation plant and the creation of AO BAT Kyrgyzstan. 4 Oct 1994. Guildford Depository. BAT. Bates No: 203477299-315.

729 A Business Development Plan proposed by BAT Industries plc for an investment in Bishkek tobacco factory and Bishkek fermentation plant and the creation of AO BAT Kyrgyzstan. 4 Oct 1994. Guildford Depository. BAT. Bates No: 203477316-39

730 World Bank. Economics of tobacco for the Europe and Central Asia (ECA) Region, May 2001. http://www1.worldbank.org/tobacco/pdf/country%20briefs/Eastern%20and%20Central %20Asia.pdf (last accessed 11/6/02)

731 Glogan T. Contraband impairs Estonian market. Tobacco Journal International 1999;5:69.

732 Collin J, Lee K, Bissell K. Negotiating the Framework Convention on Tobacco Control: The politics of global health governance. In: Wilkinson R and Murphy C eds. Global Governance: A reader London: Routledge, 2005.

733 Unknown . Draft 2, Project Umpire, Skeleton business Plan. [01/02/1994]. Guildford Depository. BAT. Bates No: 600507395-420.

734 Burton R. B&W - Central Asia. 08/12/1993. Guildford Depository. BAT. Bates No: 203459274-5.

735 Wells W. Project Umpire: Bain Link Market Sizing Model. 11/07/1994. Guildford Depository. BAT. Bates No: 503801054-5.

736 Marr M. Market sizing - Uzbekistan draft 2. 03/08/1994. Guildford Depository. BAT. Bates No: 203459203-4.

737 van Waay A. Uzbekistan - market sizing estimates. 24/06/1994. Guildford Depository. BAT. Bates No: 203459354-5.

398

738 Aitken D. Notes on Visit to Uzbekistan 6-10 June 1994. 13/06/1994. Guildford Depository. BAT. Bates No: 600507743-50.

739 Marr M, Sims D. Uzbekistan Marketing Plan 1994-2003 Discussion Draft.

[01/10/1993]. Guildford Depository. BAT. Bates No: 203459182-201.

740 Wells W. Project Umpire: issues for Dr. Khamidov et al and Revised draft BAT's proposed investment in Uzbekistan. Issues for discussion. 29/11/1993. Guildford Depository. BAT. Bates No: 301728237-41.

741 British American Tobacco. Decree No. 30. "On Restricting and Preventing Smoking in Uzbekistan". 25/08/1994. Guildford Depository. BAT. Bates No: 500898510-1.

742 Chief Sanitary Doctor of the Republic of Uzbekistan. Decree no. 30 “On restricting and preventing smoking in Uzbekistan”. 15/07/1994. Guildford Depository. BAT. Bates No. 500852350-1

743 Schroders. Project Umpire. Meeting to be held at Windsor House on 23rd August at 1430hrs Proposed agenda. 23/08/1994. Guildford Depository. BAT. Bates No: 503800858-9.

744 Herter U. Secret. Supplemental note to the Chief executive's committee. Uzbekistan. 18/10/1994. Guildford Depository. BAT. Bates No: 503800812-4.

745 Rowlands-Rees G. Meeting 27th August 1994. 27/08/1994. Guildford Depository. BAT. Bates No: 500898526-7.

746 Rowlands-Rees G. Meeting at Ministry of Health 30th August 1994. 30/08/1994. Guildford Depository. BAT. Bates No: 500898516-21.

747 Wilson M. BAT Industries PLC Reporting Minutes. Secret. 05/09/1994. Guildford Depository. BAT. Bates No: 203470848-9.

748 Boyse S. Tashkent Decree. 26/08/1994. Guildford Depository. BAT. Bates No: 500852336-9.

399

749 Smith S. [email re Uzbek deal]. 12/10/1994. Guildford Depository. BAT. Bates No: 503800830.

750 Norwell I. Uzbekistan Decree. 26/08/1994. Guildford Depository. BAT. Bates No: 500852340-3.

751 Bruce-Miller N. Sir Patrick Sheehy's Visit to Uzbekistan and the Cabinet of Minister's Decree. 13/06/1994. Guildford Depository. BAT. Bates No: 600507740-2.

752 Rowland-Rees J. Meeting at Minstry of Health with Mr Iskandarov. 30/08/1994. Guildford Depository. BAT. Bates No: 500898515.

753 Boyse S. Tashkent.doc. CONFIDENTIAL notes. [2/9/94]. Guildford

Depository. BAT. Bates No: 500852331-5.

754 British American Tobacco. Code of tobacco advertising conduct for the Russian Federation. International Tobacco companies operating in Russia. 31/12/1994. Guildford Depository. BAT. Bates No: 500852344-7.

755 Boyse S. Uzbek.doc. 02/09/1994. Guildford Depository. BAT. Bates No: 500852324-6.

756 Boyse S. Uzbek2.doc. 02/09/1994. Guildford Depository. BAT. Bates No:

500852327-30.

757 Boyse S. Uzbekistan Decree. 02/09/1994. Guildford Depository. BAT. Bates No: 500852323.

758 Wells W. Project Umpire: Current o/s position. 20/09/1994. Guildford Depository. BAT. Bates No: 600507555-61.

759 Schroders. Project Umpire: the GKI package. 12/10/1994. Guildford Depository. BAT. Bates No: 503800773-85.

760 Unknown . British-American Tobacco Company Limited Management Board Meeting. 19/9/94. Guildford Depository. BAT. Bates No: 500046165-70.

761 Unknown . ORDER of the Cabinet of Ministers of the Republic of Uzbekistan. 19/09/1994. Guildford Depository. BAT. Bates No: 500898504.

400

762 King J, Wells W. Project Umpire. 22/09/1994. Guildford Depository. BAT. Bates No: 503800853-5.

763 Wells W. Project Umpire. 07/10/1994. Guildford Depository. BAT. Bates No: 503800851-2.

764 Wells W. Urgent Project Umpire completion. 26/10/1994.

Guildford Depository. BAT. Bates No: 600507538-45.

765 Simpson D. Uzbekistan: who’s in charge now? Tobacco Control 2000; 9: 359- 361.

766 Unknown author. Draft 1. BAT/Kyrgyzstan. Guildford Depository. BAT. Bates No: 503912326-330.

767 J Henry Schroder Wagg & Co. Ltd. Preliminary review of the Transcaucasian and Central Asian Republics of the Former Soviet Union. Volume II: Full report. 28/08/1992. Guildford Depository. BAT. Bates No: 203456596-758.

768 World Health Organization (2004) Building Blocks for Tobacco Control: A Handbook (Geneva: WHO)

769 Richards JW Jr, Tye JB, Fischer PM. The tobacco industry's code of advertising in the United States: myth and reality. Tob Control 1996; 5: 295-311.

770 Jones K, Wakefield M, Turnbull DA. Attitudes and experiences of restaurateurs regarding smoking bans in Adelaide, South Australia. Tob Control 1999; 8: 62-6.

771 Szilagyi T, Chapman S. Hungry for Hungary. Examples of tobacco industry’s expansionism Cen Eur J Publ Health. 2003; 11: 38-43.

772 Szilagyi T, Chapman S. Tobacco industry efforts to keep cigarettes affordable: a case study from Hungary. Cen Eur J Publ Health. 2003; 4: 223-8.

773 Szilagyi T, Chapman S. Tobacco industry efforts to erode tobacco advertising controls in Hungary. Cen Eur J Publ Health. 2004; 12: 190-6

401

774 MacKenzie R, Collin C, Lee K. “Almost a role model of what we would like to do everywhere” British American Toabcco in Cambodia. Tobacco Control 2004; 13 supp II: ii112-7.

775 Patel P, Collin C, Gilmore A. The law was actually drafted by us but the Government is to be congratulated on its wise actions”: British American Tobacco and public policy in Kenya. (submitted) British Medical Journal.

776 Onder and Bilir,2000 quoted in Yurekli AA. De Beyer J. Did the entry of private cigarette producers in Turkey and Ukraine increase cigarette consumption? Unpublished report.

777 Nye D. Regulatory myopia and public health: “Tough” tobacco control. Competition and Change 2004; 8: 305-21

778 Soderlund N, Tangcharoensathien V. Health sector regulation – understanding responses from government. Health Policy and Planning 2000; 15: 347-8.

779 MacKenzie R, Colin J, Sriwongcharoean K, Muggli M. “If we can just ‘stall’ new unfriendly legislations, the scoreboard is already in our favour”: transnational tobacco companies and ingredients disclosure in Thailand.

780 World Health Organization. Framework Convention on Tobacco Control. 21 May 2003. http://www.who.int/tobacco/areas/framework/final_text/en/ (accessed: 6 Jul 04)

781 Bozicevic I, Gilmore A, Oreskovic S. The Tobacco Epidemic in South-East Europe: Consequences and Policy Responses. Health, Nutrition and Population Discussion Paper, Economics of Tobacco Control Paper No. 18. Washington DC: World Bank, 2004. (http://www1.worldbank.org/tobacco/pdf/Bozicevic-Consequences%20of- whole.pdf)

782 Callard C, Thompson D, Collishaw N. Transforming the tobacco market: why the supply of cigarettes should be transferred from for-profit corporations to non-profit enterprises with a public health mandate. Tobacco Control 2005; 14: 278-83.

402

783 Callard C, Thompson D, Collishaw N. Curing the addiction to profits: A supply- side approach to phasing out tobacco. Ottowa: Canadian Centre for Policy Alternatives, 2005.

403